TOY SYSTEM AND A METHOD OF OPERATING THE TOY SYSTEM

Abstract

A toy system including a plurality of toy construction elements, an image capturing device, and a processor, wherein the image capturing device is operable to capture one or more images of a toy construction model constructed from the toy construction elements. The processor is configured to execute a digital game configured to cause the processor to provide a digital play experience. The processor is configured to receive an unlock code and unlock virtual objects for use in the digital play experience. The virtual objects are associated with toy construction elements. The processor is configured to receive images captured by the image capturing device, recognize one or more toy construction elements within the images, and provide a digital play experience involving the unlocked virtual. Objects.

Claims

1. A toy system, comprising: a plurality of toy construction elements, an image capturing device and a processor, wherein the image capturing device is operable to capture one or more images of a toy construction model constructed from the toy construction elements: wherein the processor is configured to: execute a digital game, the digital game comprising computer executable code configured to cause the processor to provide a digital play experience; receive an unlock code indicative of one or more virtual objects; responsive to receiving the unlock code, unlock the one or more virtual objects associated with said received unlock code for use in the digital play experience, each virtual object being associated with a respective one of said toy construction elements or with a respective toy construction model constructed from the toy construction elements; receive one or more images captured by said image capturing device; recognize one or more toy construction elements or toy construction models with the one or more images; responsive to recognizing a first toy construction element or a first toy construction model associated with a first one of the unlocked virtual objects, provide a digital play experience involving said first unlocked virtual object.

2. A toy system according to claim 1, wherein the unlock code is provided as a physical item carrying a machine readable or human readable code.

3. A toy system according to claim 1, wherein the unlock code is of limited-use and wherein the processor is configured to determine whether the received unlock code has previously been used beyond the limited use, and to unlock the virtual object only, if the code has not previously been used beyond the limited use.

4. A toy system according to claim 1, wherein the digital game comprises computer executable code configured to cause the processor to control at least one virtual game item.

5. A toy system according to claim 1, wherein the unlocked first virtual object has a visual appearance that resembles the first toy construction element or model with which the first virtual object is associated.

6. A toy system according to claim 1, wherein the processor is configured, responsive to receiving the unlock code, to associate a visual appearance to the unlocked virtual object, in particular by receiving one or more captured images of a toy construction model whose visual appearance is to be associated with the unlocked virtual object; and to associate the visual appearance of said toy construction model with the unlocked virtual object.

7. A toy system according to claim 1, wherein the received one or more images, within which the processor recognizes one or more toy construction elements or toy construction models, depicts a composite toy construction model constructed from at least a first toy construction model and a second toy construction model; wherein recognizing one or more toy construction elements or toy construction models within the one or more images comprises recognizing each of the first and second toy construction models included in the composite toy construction model; and wherein the processor is configured, responsive to recognizing the first and second toy construction models, to provide a digital play experience involving said first and second unlocked virtual objects, wherein the first toy construction model is associated with a first unlocked virtual object and the second toy construction model is associated with a second unlocked virtual object

8. A toy system according to claim 7, wherein the processor is configured to further recognize a spatial configuration of the first and second toy construction models relative to each other, and to modify the provided play experience responsive to the recognized spatial configuration.

9. A toy system according to claim 1, wherein one or more of the plurality of toy construction elements may include a visually recognizable code identifying a toy construction element or a toy construction model.

10. A toy system according to claim 9, wherein the plurality of toy construction elements includes one or more marker toy construction elements each having a visual appearance representative of an object code or a part thereof.

11. A toy system according to claim 10, wherein the processor is configured to detect the object code within the one or more images and to adapt the digital play experience responsive to the detected object code.

12. A toy system according to claim 1, wherein recognizing the first toy construction element or the first toy construction model associated with the first unlocked virtual object comprises: recognizing the first toy construction element as a toy construction element of a first type of toy construction elements or recognizing the first toy construction model as a toy construction model of a first type of toy construction models; and detecting a first object code associated with the recognized first toy construction element or the recognized first toy construction model.

13. A toy system according to claim 12, wherein the processor is configured, responsive to recognizing the first toy construction element or the first toy construction model, to provide a digital play experience involving a first instance of a plurality of instances of said first unlocked virtual object, the first virtual object being associated with the first type of toy construction element or the first type of toy construction model, and each of the plurality of instances of said virtual object being further associated with a respective object code.

14. A toy system according to claim 13, wherein the processor is configured, responsive to recognizing the first toy construction element or the first toy construction model associated with a first one of the unlocked virtual objects, to store the detected first object code associated with the first unlocked virtual object.

15. A toy system according to claim 14, wherein the processor is configured to determine whether an object code has previously been stored associated with the first unlocked virtual object and to associate the detected first object code with the first unlocked virtual object only if no object code has previously been associated with the first unlocked virtual object.

16. A toy system according to claim 15, wherein the processor is configured to compare the detected first object code with a previously stored object code associated with the first unlocked virtual object, and to provide the digital play experience involving said first unlocked virtual object only if the detected first object code corresponds to the previously stored object code associated with the first unlocked virtual object.

17. A method, implemented by a processor, of operating a toy system, the toy system comprising a plurality of toy construction. elements, an image capturing device, and the processor, the image capturing device being operable to capture one or more images of one or more toy construction models constructed from the toy construction elements and placed within a field of view of the image capturing device, wherein the method comprising: executing a digital game, the digital game comprising computer executable code configured to cause the processor to provide a digital play experience; receiving an unlock code indicative of one or more virtual objects; responsive to receiving the unlock code, unlocking the one or more virtual objects associated with said received unlock code for use in the digital play experience, each virtual object being associated with a respective one of said toy construction elements or with a respective toy construction model constructed from the toy construction elements; receiving one or more images captured by said image capturing device; recognizing one or more toy construction elements or toy construction models within the one or more images; and responsive to recognizing a first toy construction element or a first toy construction model associated with a first one of the unlocked virtual objects, providing a digital play experience involving said first unlocked virtual object

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0209] Preferred embodiments of the disclosure will be described in more detail in connection with the appended drawings.

[0210] FIG. 1 shows the steps of creating a physical model according to one embodiment,

[0211] FIG. 2 shows the steps of creating a virtual model from the physical model created by the steps of FIG. 1.

[0212] FIGS. 3-7 show the steps of creating a virtual game environment from a physical model according to a further embodiment,

[0213] FIGS. 8-17 show the steps of installing and playing a cyclic interactive game according to a yet further embodiment.

[0214] FIG. 18 is a physical playable character model with different physical tool models.

[0215] FIG. 19 is an interactive game system inch ding a physical playable character model.

[0216] FIG. 20 is an example of a triangle indexing scheme in a mesh.

[0217] FIG. 21 is an example of a process for creating a virtual environment.

[0218] FIG. 22A is an example of a mesh representation of a virtual object.

[0219] FIG. 22B is an example of a voxel representation of the virtual object of FIG. 22A.

[0220] FIG. 23 is an example of a process for converting a mesh into a voxel representation.

[0221] FIG. 24 is an illustration of an example of a triangle and a determined intersection point X of the triangle with the voxel space.

[0222] FIG. 25 is a flow diagram of an example of a process for determining the intersection of a mesh with a voxel space.

[0223] FIG. 26 is an illustration of an example of a voxelization process.

[0224] FIG. 27 is an illustration of a color space.

[0225] FIG. 28 is another example of a process for creating a virtual environment.

[0226] FIGS. 29A-B show the steps of creating a virtual game environment from a physical model according to a further embodiment.

[0227] FIG. 30 shows the steps of creating it virtual game environment from a physical model according to a further embodiment.

[0228] FIG. 31 shows another example of a process for creating a virtual environment.

[0229] FIGS. 32-34 schematically illustrate examples of toy construction sets of a toy system described herein.

[0230] FIGS. 33-37 schematically illustrate examples of use of an embodiment of toy system described herein.

[0231] FIG. 38 shows a flow diagram of an example of a process as described herein.

[0232] FIG. 39A-C show examples of toy construction models for use with a toy system as described herein.

[0233] FIG. 40 schematically illustrates another example of a use of an embodiment of a toy system described herein.

[0234] FIGS. 41-42 schematically illustrate examples of a toy system described herein.

DETAILED DESCRIPTION

[0235] Embodiments of the method and system disclosed herein may be used in connection with a variety of toy objects and, in particular with construction toys that use modular toy construction elements based on dimensional constants, constraints and matches, with various assembly systems like magnets, studs, notches, sleeves, with or without interlocking connection etc. Examples of these systems include but are not limited to the toy constructions system available under the tradename LEGO. For example, U.S. Pat. No. 3,005,282 and USD253711S disclose one such interlocking toy construction system and toy figures, respectively.

[0236] FIG. 1 shows steps of creating a physical model according to one embodiment. The virtual part of the game is played on a mobile device 1, such as a tablet computer, a portable computer, or the like. The mobile device 1 has a capturing device, data storage, a processor, and a display. It will be appreciated, however, that the various embodiments of the process described herein may also be implemented on other types of processing devices. The processing device may comprise a capturing device, data storage, a processor, and a display integrated into a single physical entity; alternatively, one or more of the above components may be provided as one or more separate physical entities that may be communicatively connectable with each other otherwise to allow data transfer between them. Game software installed on the mobile device 1 adapts the mobile device 1 for performing the method according to one embodiment of the disclosure within the framework of an interactive game. The mobile device 1 presents a building tutorial to the user 99. Following the instructions of the tutorial, the user 99 finds a number of physical objects 2, 3, 4, 5, 6 and arranges these physical objects 2, 3, 4, 5, 6 on a physical play zone 7, such as a table top or a floor space, to form a physical model 10 of a game environment. Advantageously, the building tutorial includes hints 11 on how certain predetermined physical properties of physical objects in the physical model of the game environment will be translated by the game system into characteristics of the virtual game environment to be created. This allows the user 99 to select the physical objects 2, 3, 4, 5, 6 according to these predetermined physical properties to wilfully/intentionally build the physical model in order to create certain predetermined characteristics/a certain game behaviour of the virtual game environment according to a predetermined set of rules. By way of example, FIG. 1 shows a hint in the form of a translation table indicating how different values of a predetermined physical property, here colour, are handled by the system, in particular, the user 99 is presented with the hint that green colours on physical objects will be used to define jungle elements, red colours will be used to define lava elements, and white colours will be used to define ice elements in the virtual game environment.

[0237] FIG. 2 illustrates steps of creating a virtual model from the physical model 10 created by arranging physical objects 2, 3, 4, 5, 6 on a physical play zone 7. The mobile device 1 is moved along a scanning trajectory 12 while capturing image/scan data 13 of the physical model 10. The image data is processed by the processor of the mobile device 1 thereby generating a digital three-dimensional representation indicative of the physical model 10 as well as information on predetermined physical properties, such as colour, shape and/or linear dimensions. The digital three-dimensional representation may be represented and stored in a suitable form in the mobile device, e.g. in a mesh form. The mesh data is then converted into a virtual toy construction model using a suitable algorithm, such as a mesh-to-LXFML code conversion algorithm as further detailed below. The algorithm analyses the mesh and calculates an approximated representation of the mesh as a virtual toy construction model made of virtual toy construction elements that are direct representations of corresponding physical toy construction elements.

[0238] Referring to FIGS. 3-7, steps of creating a virtual game environment from a physical model according to a further embodiment are illustrated by means of screen shots from a mobile device used for performing the steps. FIG. 3 shows an image of a setup of different everyday items found in a home and in a children's room as seen by a camera of the mobile device. These items are the physical objects used for building the physical model of the virtual game environment to be created by arranging the items on a table. The physical objects have different shapes, sizes and colours. The items include blue and yellow sneakers, a green lid for a plastic box, a green can, a folded green cloth, a yellow box, a red pitcher, a grey toy animal with a white tail, mane and forelock as well as a red cup placed as a fez hat, and further items. In FIG. 4 the physical model is targeted using an augmented reality grid overlaid to the view captured by the camera of the mobile device. The camera is a depth sensitive camera and allows for a scaled augmented reality grid to be shown. The augmented reality grid indicates the targeted area captured, which in the present case is a square of 1m by 1m. FIG. 5 shows a screen shot of the scanning process, where the mobile device with the camera pointed at the physical model is moved around, preferably capturing image data from all sides as indicated by the arrows and the angular scale. However, a partial scan. may be sufficient depending on the nature of the three-dimensional image data required for a given virtual game environment to be created. FIG. 6 shows a screen shot after a brickification engine has converted the three-dimensional scan data into a virtual toy construction model made to scale from virtual toy construction elements. The virtual toy construction model also retains information about different colours in the physical model. In FIG. 7 the virtual toy construction model has been enhanced by defining game controlling elements into the scene, thereby creating a virtual game environment where essentially everything appears to be made of virtual toy construction elements. FIG. 7 shows a screen shot of a playable figure exploring the virtual game environment. The playable figure is indicated in the foreground as a colourless/white, three-dimensional virtual mini-figure. Buttons on the right hand edge of the screen are user interface elements for the game play.

[0239] Now referring to FIGS. 8-17, steps of installing and playing a cyclic interactive game according to a yet further embodiment are illustrated schematically. In FIG. 8, the software required for configuring and operating a mobile device for its use in an interactive game system according to the present disclosure is downloaded and installed. Upon startup of the game software, a welcome page may be presented to the user as seen in FIG. 9, from which the user enters a building mode. The user may now be presented with a building tutorial and proceed to building a physical model and creating a virtual game environment as indicated in FIGS. 10 and 11, and as already described above with reference to FIGS. 1 and 2. The physical objects used for constructing the physical model are grey pencils, a brown book, a white candle standing upright in a brown foot, a white cup (in the right hand of the user in FIG. 10) and a red soda can (in the left hand of the user on FIG. 10). Once the virtual game environment is created, the user may proceed to game play by exploring the virtual game environment created as seen in FIG. 12. Before embarking on a virtual mission in the virtual game environment, the user may make a number of choices as, such as selecting a playable character and/or tools from a number of unlocked choices (top row in FIG. 13). A number of locked choices may also be shown (bottom row in FIG. 13). FIGS. 14 and 15 show different screenshots of a playable character on different missions (grey figure with blue helmet). The playable character is equipped with a tool for harvesting resources (carrots). In FIG. 14, the playable character is merely on a collecting mission. Seen in the background of the screenshot of FIG. 14 is a lava mountain created from the red soda can in the physical model. The same virtual game environment created from the same physical model is also shown in FIG. 15, but from a different angle and at a different point in the course of the game. The lava mountain created from the red soda can is shown in the landscape on the right hand side. The white cup of the physical model has been turned into an iceberg surrounded in its vicinity by ice and snow. The game environment has now spawned monsters /adversaries that compete with the playable figure for the resources to be collected (e.g. carrots and minerals). and which may have to be defeated as a part of a mission. In FIG. 16, the user has successfully completed a mission and is rewarded, e.g, by an amount of in-game currency. The in-game currency can then be used to unlock new game features, such as tools/powers/new playable characters/game levels/modes or the like. After reward and unlocking of game features, the user may receive a new mission involving a rearrangement of the physical model, thereby initiating a new cycle of the interactive game. The cycle of a cyclic interactive game is shown schematically in FIG. 17. The game system provides a task (top) and the user creates a virtual game environment scene from physical objects (bottom right); the user plays one or more game segments in the virtual game environment/scene (bottom left); and in response to the outcome of the game play, a new cycle is initiated by the game system (back to the top).

[0240] FIG. 18 shows an example of a physical playable character model with different physical tool models. The physical playable character model is for use in an interactive game system. The playable character model may be fitted with a choice of the physical tool models. By way of example, a selection of physical tool models is shown in the bottom half of FIG. 18. Each physical tool model represents specific tools, powers and/or skills. FIG. 19 shows an interactive game system including the physical playable character model of FIG. 18. The physical playable character model may be used for playing, e.g. role playing, in the physical model created by means of the physical objects as shown in the background. By entering information about the physical playable character model and the tools with which it is equipped in the game, a corresponding virtual playable character model is created for game play in the virtual game environment as indicated on the display of the handheld mobile device in the foreground of FIG. 19 (bottom right). Note also, that on the schematic view a physical play zone has been defined by a piece of green card board on the table top. The green card board has been decorated with colour pencils to mark areas on the physical play zone that in the virtual game environment are converted into rivers with waterfalls over the edge of the virtual scene as shown schematically on the handheld mobile device in the foreground.

[0241] An important step in creating the virtual game environment is the conversion of the digital three-dimensional representation obtained from, or at least created on the basis of data received from, the capturing device into a virtual toy construction model constructed from virtual toy construction elements or into another voxel based representation. In the following an example will be described of a conversion engine adapted for performing such a conversion, in particular a conversion engine for conversion from a mesh representation into an LXFML. representation. It will be appreciated that other examples of a conversion engine may perform a conversion into another type of representation.

[0242] With the evolution of computers and computer vision it is becoming easier for computers to analyze and represent three-dimensional objects in a virtual world. Different technologies exist nowadays that facilitate the interpretation of the environment, creating three-dimensional meshes out of normal pictures obtained from normal cameras or out of depth camera information.

[0243] This means that computers, smartphones, tablets and other devices will increasingly be able to represent real objects inside an animated world as three-dimensional meshes. In order to provide an immersive game experience or other types of virtual game experiences, it would be of great value if whatever a computer could “see” and represent as a mesh could then be transformed into a model built out of toy construction elements such as those available under the name LEGO or at least as a voxel-based representation.

[0244] Virtual toy construction models may be represented in a digital representation identifying which virtual toy construction elements are comprised in the model, their respective positions and orientations and, optionally, how they are interconnected with each other. For example, the so-called LXFML format is a digital representation suitable for describing models constructed from virtual counterparts of construction elements available under the name. It is thus desirable to provide an efficient process for converting a digital mesh representation into a LEGO model in LXFML format or into a virtual construction model represented in another suitable digital representation.

[0245] Usually, three-dimensional models are represented as meshes. These meshes are typically collections of colored triangles defined by the corners of the triangles (also referred to as vertices) and an order of how these corners should be grouped to form these triangles (triangle indexes). There is other information that a mesh could store but the only other thing relevant for this algorithm is the mesh color.

[0246] As described earlier, the algorithm receives, as an input, mesh information representing, one or more objects. The mesh information comprises: [0247] Mesh vertices/vertex positions: the coordinates of the points that form the triangles, meaning points in space, e.g. represented as vectors (x,y,z), where x,y,z can he any real number. [0248] Triangle indexes: the indexes of the vertices in consecutive order so that they form triangles, i.e. the order in which to choose the vertices from the positions in order to draw the triangles in the mesh. For example, FIG. 20 illustrates an example of an indexing scheme for a simple surface defined by 4 points labelled 0, 1, 2 and 3, respectively, defining a rectangle. In this example, an array of indexes like {0,1,2,3,0} may be defined to represent how triangles may be defined to represent the surface. This means that a process starts from point 0, proceed to point 1, then to point 2. That is the first triangle. The process may then proceed from the current point (point 2) to define the next triangle, so the process only needs the remaining 2 points, which are 3 and 0. This is done in order to use less data to represent the triangles. [0249] Mesh color .information: the colors that the triangles have.

[0250] Embodiments of the process create a representation of a virtual construction model, e.g. an LXFML string format version 5 or above. The LXFML representation needs to include the minimum information that would be needed by other software tools in order to load the information inside. The following example will be used to explain an example of the information included n an LXFML file:

TABLE-US-00001  1 <?xml version=“1.0” encoding=“UTF-8” standalone=“no” ?>  2 <LXFML versionMajor=“5” versionMinor=“0” name=“Untitled”>  3 <Meta>  4  <Application name=“VoxelBrickExporter” versionMajor=“0” versionMinor=“1”/>  5 </Meta>  6 <Bricks>  7  <Brick refID=“0” designID=“3622”>  8   <Part refID=“0” designID=“3622” materials=“316”>  9    <Bone refID=“0” transformation=“5.960464E−08,0,0.9999999,0,1,0,−0.9999999,0,5.960464E− 08,0,1.6,20”> 10    </Bone> 11   </Part> 12  </Brick> 13 </Bricks> 14 </LXFML>

[0251] The first line merely states the format of the file.

[0252] The second line contains information about the LXFML version and the model na The LXFML version should preferably be 5 or higher. The model name serves as information only. it does not of the loading/saving process in any way.

[0253] A <Meta>section is where optional information is stored. Different applications can store different information in this section if they need to. The information stored here does not affect the loading process.

[0254] Line 4 provides optional information about what application exported the LXFML file. This may be useful for debugging purposes.

[0255] The subsequent lines include the information. about the actual toy construction elements (also referred to as bricks). The refID should be different for every brick of the model (a number that is incremented. every time a brick is exported will do just fine). The design ID gives information about the geometry of the brick and the materials give information about the color. The transformation is the position and rotation of the brick represented by a 4 by 4 matrix but missing the 3.sup.rd column.

[0256] This information is considered sufficient. One could test the validity of an LXFML file by loading it with the free tool. LEGO Digital. Designer that can be found at http://ldd.lego.com.

[0257] FIG. 21 shows a flow diagram illustrating the steps of an example of a process for converting a mesh into a representation of a virtual toy construction model. These steps are made independent because sometimes not all of them are used, depending on the situation.

[0258] In initial step S1, the process receives a mesh representation of one or more objects. For the purpose of the present description, it will be assumed that the process receives a mesh including the following information: [0259] Vm=mesh vertices. [0260] Tm=mesh triangles. [0261] Cm=mesh color; (Per vertex color)

[0262] It will be appreciated that, instead of a mesh color, the mesh may represent another suitable attribute, such as a material or the like. Nevertheless, for simplicity of the following description, reference will be made to colors.

[0263] In an initial conversion step S2, the process converts the mesh into voxel space. The task addressed by this sub-process may be regarded as the assignment of colors (in this example colors from a limited palette 2101 of available colors, i.e. colors from a finite, discrete set of colors) to the voxels of a voxel space based on. a colored mesh The mesh should fit the voxel space and the shell that is represented by the mesh should intersect different voxels. The intersecting voxels should be assigned the closest color from the palette that corresponds to the local mesh color. As this technology is used in computer implemented applications such as gaming, performance is very important.

[0264] The initial sub-process receives as an input a mesh that has color information per vertex associated with it. It will be appreciated that color may be represented in different ways, e.g. as material definitions attached to the mesh. Colors or materials may be defined in a suitable software engine for three-dimensional modelling, e.g. the system available under the name “Unity”.

[0265] The mesh-to-voxel conversion process outputs a suitable representation of a voxel space, e.g. as a three-dimensional array of integer numbers, where each element of the array represents a voxel and where the numbers represent the color ID, material ID or other suitable attribute to be assigned to the respective voxels. All the numbers should be 0 (or another suitable default value) if the voxel should not be considered an intersection; otherwise, the number should represent a valid color (or other attribute) ID from the predetermined color/material palette, if a triangle intersects the voxel space at the corresponding voxel. The valid color should preferably be the closest color from the predetermined palette to the one the triangle intersecting the voxel has.

[0266] FIG. 22A shows an example of a mesh representation of an object while FIG. 22B shows an example of a voxel representation of the same object where the voxel representation has been obtained by an example of the process described in the following with reference to FIG. 23.

[0267] So the task to be performed by the initial sub-process may be regarded as: given a mesh model, determine a voxel representation that encapsulates the mesh model and has as voxel color the closest one of a predetermined set of discrete colors to the mesh intersecting the voxel(s).

[0268] Initially converting the mesh into a voxel representation is useful as it subsequently facilitates the calculation of where different toy construction elements should be positioned, Voxels may be considered boxes of size X by Y by Z (although other types of voxels may be used). Voxels may be interpreted as three-dimensional pixels. The conversion into voxels may be useful in many situations, e.g. when the model is to be represented as virtual toy construction elements in the form of box shaped bricks of size X′ by Y′ by Z′. This means that any of the bricks that we might have in the model will take up space equal to a multiple of X, Y, Z by the world axis x, y and z.

[0269] In order to create the voxel space needed for the model that is to be converted, the process starts at step S2301 by creating an axis-aligned bounding box around the model. The bounds can be computed from the mesh information. This can be done in many ways; for example the Unity system provides a way to calculate bounds for meshes. Alternatively, a bound can he created out of two points: one containing the minimum coordinates by x, y and z of all the vertices in all the meshes and the other containing the maximum values by x, y and z, like in the following example:

Pmin.sub.x=Min.sub.x(Vm1.sub.x, Vm2.sub.x . . . ) Pmax.sub.x=Max.sub.x(Vm1, Vm2.sub.x. . . )

Pmin.sub.y=Min.sub.y(Vm1.sub.y, Vm2.sub.y . . . ) Pmax.sub.y=Max.sub.y(Vm1.sub.y, Vm2.sub.y . . . )

Pmin.sub.z=Min.sub.z(Vm1.sub.z, Vm2.sub.z . . . ) Pmax.sub.z=Max.sub.z(Vm1.sub.z, Vm2.sub.z . . . )

Pmin=(Pmin.sub.x, Pmin.sub.y, Pmin.sub.z) Pmax=(Pmax.sub.x, Pmax.sub.y, Pmax.sub.z)

[0270] Pmin and Pmax are the minimum and maximum points with coordinates x, y and z. Max and Min are the functions that get the minimum and maximum values from an array of vectors Vm by a specific axis (x,y or z).

[0271] Having the opposite corners of a box should be sufficient to define it. The box will have the size B=(bx, by, bz) by axis x, y and z. This means that B=Pmax−Pmin;

[0272] In a subsequent step S2302, the process divides he voxel space into voxels of a suitable size, e.g. a size (dimx, dimy, dimz) matching the smallest virtual toy construction element of a system of virtual toy construction elements. Preferably the remaining virtual toy construction elements have dimensions corresponding to integer multiples of the dimensions of the smallest virtual toy construction element. In one example, a voxel has dimensions (dimx, dimy, dimz)=(0.8, 0.32, 0.8) by (x,y,z) which is the size of a 1×1 Plate LEGO plate (LEGO Design ID: 3024). By creating the Voxel Space corresponding to the bounding box we will create a Matrix of size V(vx,yy,vz), where vx=bx/dimx+1, vy=by/dimy+1 and vz=bz/dmiz+1. The +1 appears because the division will almost never be exact and any remainder would result in the need of having another voxel that will need filling.

[0273] The matrix will contain suitable color IDs or other attribute IDs. This means that a voxel will start with a default value of 0 meaning that in that space there is no color. As soon as that voxel needs to be colored, that specific color ID is stored into the array. In order to process the mesh, the process processes one triangle at a time and determines the voxel colors accordingly, e.g. by performing the following steps: [0274] Step S2303: Get next triangle [0275] Step S2304: Get the intersection of the triangle with the voxel space Step S2305: Compute a raw voxel color of the intersecting voxel(s). [0276] S2306: Get the color ID of the closest color from the raw voxel color and so the subsequent bricks can be created with a valid color. Step S2307: Mark the voxels) with the determined color ID.

[0277] These steps are repeated until all triangles are processed.

[0278] The computation. of the raw voxel color to be assigned to the intersecting voxels (Step S2305) may be performed in different ways. Given the input, the color of a voxel can be calculated based on the intersection with the voxel and the point/area of the triangle that intersects with the voxel or, in case of triangles that are small and the color variation is not that big, it could be assumed that the triangle has the same color and that is the average of the 3 colors in the corners. Moreover, it is computationally very cheap to calculate the average triangle color and approximate just that one color to one of the set of target colors. Accordingly: [0279] In one embodiment, the process may simply average out the color of the triangle using the 3 vertex colors and then use the average color fir all intersections. [0280] an alternative embodiment, the process computes where on the triangle is the intersection with the voxel space. FIG. 24 illustrates an example of a triangle and a determined intersection point X of the triangle with the voxel space. Then the process computes the color as follows (see FIG. 24): [0281] Having the 3 colors Ca. Cb and Cc associated to the vertices/vectors A, B and C, the intersecting color is C=⅙ * Σ.sub.B.sup.A[(Xab * Cb+(AB−Xab) * Ca)/AB+(Xcb * Cb+(CB−Xcb) * Cc)/CB], where Xab is the distance from the intersection of the triangle with the voxel space along the AB axis, AB is the distance from a to B and the sum represents the same process applied for A, B and C to obtain a color blend.

[0282] While the first alternative is faster, the second alternative provides higher quality results.

[0283] The determination of the intersection of the triangle with the voxel space (step S2304 may be efficiently performed by the process illustrated in FIGS. 25 and 26 and as described as follows. In particular, FIG. 25 illustrates a flow diagram of an example of the process and FIG. 26 illustrates an example of a triangle 2601. Since it is fairly easy to convert a point in space to a coordinate in voxel space, the process to fill the voxels may be based on points on the triangle as illustrated in FIG. 26. The steps of the sub-process, which is performed for each triangle of the mesh, may be summarized as follows:

[0284] Step S2501: select one corner (corner A in the example of FIG. 26) and the edge opposite the selected corner (edge BC in the example of FIG. 26).

[0285] Step S2502: Define a sequence of points BC1-BC5 that divide the opposite edge (BC) into divisions equal to the smallest dimension of a voxel, e.g. dimy=0.32 in the above example. The points may be defined as end points of a sequence of vectors along edge BC, where each vector has a length equal to the smallest voxel dimension. Since, in three-dimensional, it is highly unlikely to have integer divisions, the last vector will likely end between. B and C rather than coincide with C.

[0286] The process then processes all points BC1-BC5 defined in the previous step by performing the following steps:

[0287] Step S2503: Get next point

[0288] Step S2504: The process defines a line connecting the corner picked at step S2501 with the current point on the opposite edge.

[0289] Step S2505: The process divides the connecting line into divisions with the size equal to the smallest dimension of a voxel, again dimy=0.32 in the above example. Hence, every point generated by the split of step S2502, connected with the opposite corner of the triangle (A in the example of FIG. 26) forms a line which is to he split in the same way, but starting from the point on the edge (BC) so that the last point that might not fall into the point set because of the non integer division be A. At last, AC should be split into points.

[0290] Step S2506: For every point on the line that was divided at Step S2505 and for point A, the process marks the voxel of the voxel space that contains this point with the raw color computed as described above with reference to step S2305 of FIG. 23. This mapping may be done very efficiently by aligning the model to the voxel space and by dividing the vertex coordinates to the voxel size The process may allow overriding of voxels. Alternatively, the process may compute weights to determine which triangle intersects a voxel most. However, such a computation is computationally more expensive.

[0291] This simple pseudocode shows how the voxel representation of the mesh can be created using just the mesh information. The amount of operations done is not minimal as the points towards the selected corner (selected at Step S2501) tend to be very close to each other and not all of them would be needed. Also the fact that the triangle could be turned at a specific angle would mean that the division done at Step S2506 may take more steps than necessary. However, even though there is a lot of redundancy, the operation is remarkably fast on any device and the complexity of calculations needed to determine the minimum set of points would likely result in having a slower algorithm.

[0292] Again referring to FIG. 23, the determination of the closest color from a palette of discrete colors (step S2306) may also be performed in different ways: [0293] In one embodiment, which results to a high quality color mapping, the process initially transforms RGB colors into LAB space. With the LAB representation of colors, the process applies the DeltaE color distance algorithm to compute the distance from the actual color (as determined from the mesh) and the other available colors from the palette. A more detailed. description of this method is available at http://en.wikipedia.org/wiki/Color_difference. [0294] In another embodiment, which is faster than the first embodiment, the process calculates the difference between the valid colors of the palette and the given color (as determined from the mesh). The process then selects the color of the palette that corresponds to the shortest distance.

[0295] One way to find the shortest distance is to compare all distances in three-dimensional. This means that any color that is to be approximated has to be compared with all possible colors from the palette. A more efficient process for determining the closest distance between a color and the colors of a palette of discrete colors will be described with reference to FIG. 27:

[0296] All colors in RGB space may be represented in three-dimensional as an 8.sup.th of a sphere/ball sectioned by the X, Y and Z planes with a radius of 255. If a color C with components rC, gC, bC containing the red, green and blue components is given as input for the conversion step, color C will be situated at distance D from the origin.

[0297] The minimum distance may then be found by an iterative process starting from an initial value of the minimum distance. The maximum distance from the origin to the closest target color from the palette that should be found must be no larger than the distance from the origin to the original color plus the current minimum distance. The initial minimum is thus selected large enough to cover all possible target colors to ensure that at least one match is found.

[0298] An example of how the process works is as fallows: a current minimum distance is found, meaning that there is a target color that is close to the input color. Now, no target color can be found in such way that it is closer to the original color, yet further away from origin than the distance between the original color and the origin plus the current minimum distance. This follows from the fact that the minimum distance determines the radius of the sphere that has the original color in its center and contains all possible, better solutions. Any better solution should thus be found within said sphere; otherwise it would be further away from the original color. Consequently, for a given current minimum distance, only colors need to be analyzed that are at a distance from the origin smaller than the original color distance + the current minimum.

[0299] The above conversion process results in a voxel model of the hull/contour of the object or objects. It has been found that the process provides a quality output at an astounding speed because: [0300] if all the units are at maximum distance equal to the minimum size of a voxel, one can't get 2 points that are further away than a voxel so there will never be holes. [0301] if the triangles are small and many, and if the model is big, all the small voxel overrides that might give a voxel that does not have the best color for a few voxel will be tolerable. [0302] The color approximation is good. enough while at the same time saves a lot of computation power.

[0303] This solution may be compared in performance to the standard solutions (raycasting and volume intersection) which instead of just using a given set of points in space try to determine if triangles intersect different volumes of space and, in some cases, some methods even try to calculate the points where the triangle edges intersect the voxels. The volume intersection method is expected to be the slowest, but the intersection points are expected to provide accurate areas of intersection which could potentially facilitate a slightly more accurate coloring of the voxels.

[0304] Instead of computing different intersections, another method that is commonly used to determine intersections is called raycasting. Rays can be casted in a grid to determine what mesh is hit by specific rays. The raycasting method is not only slower but also loses a bit of quality as only the triangles hit by the rays contribute to the coloring. The raycasting method could give information about depth and could help more if operations need to be done taking in the consideration the interior of the model.

[0305] Again referring to FIG. 21, the mesh-to-voxel conversion of step S2 typically results in a hollow hull, as only voxels intersecting the surface mesh have been marked with colors. In some situations it may be desirable to also map colors onto internal voxels while, in other situations, it may be desirable not to fill out the voxel model. For example, sometimes the model should be empty, e.g. when the model represents a hollow object, e.g. a ball. Moreover, it takes more time to calculate what is the inside volume of the model and it also affects the amount of bricks in the model. This makes all the subsequent steps slower because more information is handled. On the other hand, sometimes, especially if creating landscapes, it is desirable that a model is full rather than just an empty shell.

[0306] Accordingly, in the subsequent, optional step S3, the process may fill the internal, non-surface voxels with color information. The main challenge faced when trying to fill the model is that it is generally hard to detect if the voxel that should be filled is inside the model or outside. Ray casting in the voxel world may not always provide a desirable result, because if a voxel ray intersects 2 voxels, this does not mean that all voxels between the two intersection points are inside the model. If the 2 voxels contained, for example very thin triangles, the same voxel could represent both an exit and an entrance.

[0307] Raycasting on the mesh can be computationally rather expensive and sometime inaccurate, or it could be accurate but even more expensive, and therefore a voxel based solution may be used for better performance.

[0308] It is considerably easier to calculate the outside surface of the model because the process may start with the boundaries of the voxel world. If those points are all taken then everything else is inside. For every voxel that is not occupied because of triangle intersections one can start marking every point that is connected to that point as being a point in the exterior. This procedure can continue recursively and. it can fill the entire exterior of the model.

[0309] Now that the edge is marked and the exterior is marked, everything in the voxel space that is unmarked (still holds a value of 0) is inside the model.

[0310] Now, a voxel raycasting can be done to shoat rays by any axis and fill in any unoccupied voxel. Currently, the color of voxel that intersects the entering ray is used to color the interior. As the mesh holds no information about how should the interior be colored, this coloring could be changed to be application specific.

[0311] In subsequent, optional step S4, the created voxel :representation may be post-processed, e.g. trimmed. For example, such a post-processing may be desirable in order to make the voxel representation more suitable for conversion into a virtual toy construction model. For example, toy construction elements of the type known as LEGO often have coupling knobs. When the volume defined by the mesh is not too big, an extra knob could make a huge difference for the overall appearance of the model; therefore, for bodies with volumes less than a certain volume, an extra trimming process may be used. For example, the minimum volume may be selected as 1000 voxels or another suitable limit.

[0312] The trimming process removes the voxel on top of another voxel; if there is only one voxel that exists freely it is removed also. This is clone because the LEGO brick also has knobs that connect to other bricks. Since the knob of the last brick on top is sticking out it could mark another voxel but we might not want to put a brick there because it will make the already existing small model even more cluttered. For this reason the extra trimming process may optionally be used for small models. Of course, it could also be used on bigger models but it will introduce extra operations that might not provide observable results.

[0313] The trimming process may e.g.. be performed as follows: For every occupied voxel, the process checks if there is an occupied voxel on top; if not, it marks the occupied voxel for deletion. Either lonely voxels or the top most voxels will be removed this way. The voxels on top are collected and removed all at the same time because if they would remove themselves first the voxel underneath might appear as the top-most voxel.

[0314] After the voxel space is filled (and, optionally, trimmed), either just the contour or the interior also, some embodiments of the process may create a virtual environment directly based on the voxel representation while other embodiments may create a toy construction model as described herein.

[0315] Accordingly, in the subsequent step SS, the process parses the voxel space and creates a data structure, e.g. a list, of bricks (or of other types toy construction elements). It will be appreciated that, if a raw voxel representation of a virtual environment is desired, alternative embodiments of the process may skip this step.

[0316] In order to obtain the bricks that can be placed, a brick evolution model is used, i.e. a process that starts with a smallest possible brick (the 3024, 1×1 plate in the above example) and seeks to fit larger bricks starting from the same position. Hence the initial smallest possible brick is caused to evolve into other types of bricks. This can be done recursively based on a hierarchy of brick types (or other types of toy construction elements). Different bricks are chosen to evolve into specific other bricks. To this end the process may represent the possible evolution paths by a tree structure. When placing a brick the process will try to evolve the brick until it cannot evolve anymore because there is no other brick it can evolve into or because there are no voxels with the same color it can evolve over.

[0317] An example of this would be: a 1×1 Plate is placed in the origin. It will try to evolve into a 1×1. Brick by looking to see if there are 2 voxels above it that have the same color. Assuming there is only one and therefore it cannot evolve in that direction, the process will then try to evolve the brick into a 1×2 Plate in any of the 2 positions (normal, 90 degree rotated around the UP axis). If the brick is found to be able to evolve into a 1×2 plate then the process will continue until it will run out of space or evolution possibilities. In one embodiment, the supported shapes are 1×1 Plate, 1×2 Plate, 1×3 Plate, 1×1, Brick, 1×2 Brick, 1×3 Brick, 2×2 Plate, 2×2 Brick, but more or other shapes can he introduced in alternative embodiments.

[0318] After the brick evolution of a brick has finished, the process clears the voxel space at the location occupied by the evolved brick. This is done in order to avoid placing other bricks at that location. The process then adds the evolved. brick to a brick list.

[0319] The list of bricks thus obtained. contains information about how to represent the bricks in a digital world with digital colors.

[0320] Optionally, in subsequent step S6, the process modifies the created toy construction model, e.g. by changing attributes, adding game-controlling elements and/or the like as described herein. This conversion may be at least in part be performed based on detected physical properties of the real world scene, e.g. as described above.

[0321] In subsequent step S7, the process creates a suitable output data structure representing the toy construction model. For example, in one embodiment, the bricks may be converted into bricks that are suitable to he expressed as an LXFML file.

[0322] This means that a transformation matrix may need to be calculated and, optionally, the colors may need to be converted to a valid color selected from a predetermined color palette (if not already done in the previous steps).

[0323] The transform matrix may be built to contain the rotation as a quaternion, the position and the scale (see e.g.

[0324] http://www.euclideanspace.com/maths/geometry/affine/matrix4×4/ for more detailed information on matrices and

[0325] http://www.euclideanspace.com/maths/geometry/rotations/conversions/quaternionTo Matrix/ more info on quaternion transformation). All the bricks may finally be written in a suitable data format, e.g. in the way described above for the case of an LXMF format.

[0326] With reference to FIGS. 28. 29A-13 and 30, another embodiment of a process for creating a virtual game environment from a physical model will now be described. In particular, FIG. 28 shows a flow diagram of another embodiment of a process for creating a virtual game environment from a physical model and FIGS. 29A-B and 30 illustrate examples of steps of creating a virtual game environment from a physical model according to a further embodiment.

[0327] In initial step S2801, the process obtains scan data, i.e. a digital three-dimensional representation of the physical model, e.g. as obtained by scanning the physical model by means of a camera or other capturing device as described herein. The digital three-dimensional representation may be in the form of a surface mesh as described. herein. FIG. 29A illustrates an example of a scanning step for creating a virtual model from a physical model of a scene. The physical model of the scene comprises physical objects 2902 and 2903 arranged on a table or similar play zone. A mobile device 2901 is moved along a scanning trajectory while capturing image/scan data of the physical model. In this example, the physical objects include a number of everyday objects 2902 and a physical toy construction. model 2903 of a car.

[0328] In step S2802, the process recognizes one or more physical objects as known physical objects. To this end, the process has access to a library 2801 of known physical objects, e.g. a database including digital three-dimensional representations of each known object and, optionally, additional information such as attributes to be assigned to the virtual versions of these objects, such as functional attributes, behavioral attributes, capabilities, etc. In the example of FIGS. 29A-B, the process recognizes the physical toy construction model 2903 as a known toy construction model.

[0329] In step S2803, the process removes the triangles (or other geometry elements) from the mesh that correspond to the recognized object, thus creating a hole in the surface mesh.

[0330] In step S2804, the process fills the created hole by creating triangles filling the hole. The shape and colors represented by the created triangles may be determined by interpolating the surface surrounding the hole. Alternatively, the created surface may represent colors simulating a shadow or after-glow of the removed object.

[0331] In subsequent step S2805. the process creates a virtual environment based on the thus modified mesh, e.g. by performing the process of FIG. 21.

[0332] In subsequent step S2806, the process creates a virtual object based on the information retrieved from the library of know objects. For example, the virtual object may be created as a digital three-dimensional representation of a toy construction model. The virtual object may then be inserted into the created virtual environment at the location where the mesh has been modified, i.e. at the location where the object had been recognized. The virtual object is thus not merely a part of the created landscape or environment but a virtual object (e.g, a virtual item or character) that may move about the virtual environment and/or otherwise interact with the created environment. FIG. 29B illustrates an example of the created virtual environment where the physical objects 2902 of the real-world scene are represented by a virtual toy construction model 2912 as described herein. Additionally, a virtual Object 2913 representing the recognized car is placed in the virtual environment as a user-controllable virtual object that may move about the virtual environment in response to user inputs. The virtual environment of FIG. 29 is stored on the mobile device or on a remote system, e.g. in the cloud so as to allow the user to engage in digital game play using the virtual environment even when the user is no longer in the vicinity of the physical model or when the physical model no longer exists. It will be appreciated that the process may also be performed in an augmented reality context, where the virtual environment is displayed in real time while the user captures images of the physical model, e.g. as illustrated in FIG. 30.

[0333] FIG. 31 shows a flow diagram of another embodiment of a process for creating a virtual game environment from a physical model.

[0334] In initial step S3101, the process obtains scan data, i.e. a digital three-dimensional representation of the physical model, e.g. as obtained by scanning the physical model by means of a camera or other capturing device as described herein. The digital three-dimensional representation may be in the form of a surface mesh as described herein.

[0335] In step S3102, the process recognizes one or more physical objects as known physical objects. To this end, the process has access to a library 3101 of known physical objects, e.g. a database including information such as information about a predetermined theme or conversion rules that are associated with and should be triggered by the recognized object.

[0336] In subsequent step S3103 the process creates a virtual environment based on the thus modified mesh, e.g. by performing the process of FIG. 21.

[0337] In subsequent step S3104, the process modifies the created virtual environment by applying one or more conversion rules determined from the library and associated with the recognized object.

[0338] It will be appreciated that, in some embodiments, the process may, responsive to recognizing a physical object, both modify the virtual environment as described in connection with FIG. 31 and replace the recognized object by a corresponding virtual object as described m connection with FIG. 28.

[0339] Embodiments of the method described herein can be implemented by means of hardware comprising several distinct elements, and/or at least in part by means of a suitably programmed microprocessor.

[0340] Turning now to FIG. 32, an example of a toy construction set of a toy system described herein is schematically illustrated. The toy construction set is obtained in a box 4110 or other form of packaging,. The box includes a plurality of conventional toy construction elements 4120 from which one or more (two in the example of FIG. 32) toy construction models 4131, 4132 can be constructed. In the example of FIG. 32, a toy figurine 4131 and a toy car 4132 can be constructed from the toy construction elements of the set. The toy construction set further comprises two cards 4141, 4142, e.g, made from plastic or cardboard. Each card shows an image or other representation of one of the toy construction models and a machine readable code 4171, 4172, in this example a QR code, which represents an unlock code for a virtual object associated with the respective toy construction model, e.g. a virtual character and a virtual car, respectively. Alternatively, the unlock code(s) may be provided to the user in a different manner, e.g. by mail or sold separately. Hence, each unlock code is a unique code that comes with the product or is given to the user, in the form of a physical printed code or a digital code. The unlock code(s) when used (e.g. scanned or typed in) then unlocks the possibility of using, computer vision to select the unlocked virtual object in/for a digital experience.

[0341] FIG. 33 schematically illustrates another example of a toy construction set, similar to the set of FIG. 1 in that the toy construction set is obtained in a box 4110 and that the box includes a plurality of conventional toy construction elements 4120 from which one or more (two in the example of FIG. 33) toy construction models 4131, 4132 can be constructed. However, in this example, the toy construction set only includes a single card 4141 with an unlock code 4171 associated with one of the toy construction models (in this example figurine 4131) that can be constructed front the toy construction elements of the set. It will be appreciated that, in other embodiments, the set may include one or more unlock codes for unlocking one or more virtual objects associated with any subset of toy construction models constructible from the toy construction elements of the set.

[0342] FIG. 34 schematically illustrates another example of a toy construction set, similar to the set of FIG. 32. However, in this example, the toy construction set includes toy construction elements (not explicitly shown) for constructing the figurine 4131 and the car 4132 and an additional toy construction element 4123, in this example a sword 4123 that can be carried by the figurine 4131, i.e. that can be attached to a hand 4178 of the figurine 4131.

[0343] The set may include three cards 4141-4143 with respective unlock codes 4171-4173, one code 4171 associated with the figurine 4131, another code 4172 associated with the car 4132 and yet another code 4173 associated with the sword 4123, ln an alternative embodiment, the set may include a card 4144 with a single unlock 4174 code for unlocking multiple virtual objects associated with the figurine, the car and the sword, respectively. Again, it will be appreciated that, instead of providing a card 4144, the single unlock code may be provided in a different manner.

[0344] FIG. 35 illustrates an example of a use of the toy system described herein, using the toy construction set of any of FIGS. 32-34 and a suitably programmed portable device 4450, e.g. a tablet or smartphone executing an app that implements a digital game of the toy system. As in the previous examples, the toy construction set includes toy construction elements (not explicitly shown for constructing a figurine 4131 and a car 4132.

[0345] Initially, the processing device reads the one or more unlock codes included in the toy construction set, e.g. from respective cards 4141 and 4142 as described above. This causes the corresponding virtual objects 4451, 4452 (in this example a virtual car 4452 and a virtual character 4451) to be unlocked. The user may then capture an image of the toy figurine 4131 positioned in the driver's seat of the toy car 41.32. The processing device 4450 recognises the figurine and the car making up the thus constructed composite model causing the digital game to provide a play experience involving a virtual car 4452 driven by a corresponding virtual character 4451. After completion of the play experience, the user may capture another image of the same or of a different toy construction model and engage in the same or a different play experience involving the corresponding unlocked virtual objects.

[0346] Hence, while a virtual object may only need to be unlocked once it may, once unlocked, be available multiple times (e.g. a limited number or an unlimited number of times) for selection as a part of a play experience. The selection is performed by capturing an image of the corresponding physical toy construction element or model.

[0347] FIG. 36 illustrates an example of another use of the toy system described herein, e.g. using the toy construction set of any of FIGS. 32-34 and a suitably programmed portable device 4450, e.g, a tablet or smartphone executing an app that implements a digital game of the toy system. The example of FIG. 36 is similar to the example of FIG. 35. However, while the use of FIG. 35 allows the virtual objects 4451, 4452 to repeatedly be selected, optionally in different combinations with other objects, the use of FIG. 36 only allows a single selection of an unlocked virtual object. Once a combination is selected, it is the thus selected combination that is used in the play experience.

[0348] FIG. 37 illustrates an example of another use of the toy system described herein, e.g. using the toy construction set of any of FIGS. 32-34 and a suitably programmed portable device 4450, e.g. a tablet or smartphone executing an app that implements a digital game of the toy system. The example of FIG. 37 is similar to the example of FIG. 35. In particular, the toy system of FIG. 37 comprises toy construction elements from which a number of toy construction models 4131-4134 can be constructed. The toy system further comprises four cards 4141-4144 with unlock codes for unlocking four virtual objects 4451-4454, each corresponding to one of the toy construction models 4131-4134.

[0349] In the example of FIG. 37 the user has unlocked four virtual objects 4451-4454 that can be combined in different ways in the digital game by capturing images of corresponding composite toy construction models 4661-4664 constructed from respective combinations of the individually recognizable toy construction models. In particular, composite toy construction model 4661 is constructed from figurine 4131 and car 4132, composite toy construction model 4662 is constructed from figurine 4131 and car 4133, composite toy construction model 4663 is constructed from figurine 4134 and car 4132, and composite toy construction model 4664 is constructed from figurine 4134 and car 4133.

[0350] FIG. 38 shows a flow diagram of an example of a computer implemented process for controlling a digital game of a toy system, e.g, of any of the toy systems described in connection with FIGS. 32-37. In particular, the process may be executed by a processing device including a digital camera and a display, such as a mobile phone, a tablet computer or another personal computing device.

[0351] In initial step S4101, the process initiates execution of a digital game, e.g. by executing a computer program stored on a processing device. The digital game provides functionality for acquiring unlock codes, capturing images of toy construction models, recognizing toy construction models in the captured images, and for providing a digital play experience involving one or more virtual objects.

[0352] In step S4102. the process acquires an unlock code, e.g. by reading a QR code, reading an RFID tag, receiving a code manually entered by a user input, or in another suitable way.

[0353] In subsequent step S4103, the process unlocks a virtual object associated with the received unlock code. For example, the digital game may have stored information about a plurality of virtual objects, each virtual object having associated with it a stored unlock code or a set of unlock codes. The process may thus compared the acquired unlock code with the stored. unlock codes or codes so as to identify which virtual object to unlock. The process may then flag the virtual object as unlocked. In some embodiments, the process may be implemented by a distributed system, e.g. including a client device and a remote host system, e.g. as described in connection with FIG. 42. In such a system, the processing device may forward the acquired unlock code to the host system and the host system may respond with information about a virtual object to be unlocked.

[0354] In step S4104, the process receives an image of a toy construction model. For example, the image may be an image captured by a digital camera of the device executing the process. The image may directly be forwarded from the camera to the recognition process. To this end, the process may instruct the user to capture an image of a toy construction model constructed by the user, where the toy construction model represents the unlocked virtual object. In some embodiments, the process may initially display or otherwise present building instructions instructing the user to construct a predetermined toy construction model. The process may receive a single captured image or a plurality of images, such as a video stream, e.g. a live video stream currently being captured by the camera.

[0355] In step S4105, the process processes the received image in an attempt to recognize a known toy construction model in the received image. For example, the process may feed the captured image into a trained machine learning algorithm, e.g. a trained neural network, trained to recognize each of a plurality of target toy construction models. An example of a process for recognizing toy construction models is described in WO 2016/075081. However, it will be appreciated that other image processing and vision technology techniques may be used for recognizing toy construction models in the received image. It will further be appreciated that the recognition process may recognize the toy construction model as a whole or the process may recognize individual toy construction elements of the model, e.g. one or more marker toy construction elements comprising a visual marker indicative of the toy construction model.

[0356] If the process fails to recognize a known toy construction model, the process may repeat step S4105 to receive a new image. Repeated failure to recognize a known toy construction model may cause the process to terminate or to proceed in another suitable manner, e.g. requesting the user to capture another image of another toy construction model.

[0357] When the process has recognized a known toy construction model in the received image, the process proceeds at step S4106 where the process determines whether an unlocked virtual object is associated with the recognized toy construction model. To this end, the process may compare the recognized toy construction model with a list of known toy construction models, each known toy construction model having a respective virtual object associated with it. Moreover, each virtual object may have a locked/unlocked flag associated with it. Hence, only when the recognized toy construction model has a virtual model associated with it where the unlock flag is set, the process determines that an unlocked virtual object is associated with the recognized toy construction model.

[0358] When the process determines that an unlocked virtual object is associated with the recognized toy construction model, the process proceeds at step S4107. Otherwise, the process may terminate, inform the user that the corresponding virtual object needs to be unlocked, or proceed in another suitable manner.

[0359] At step S4107, the process provides a digital play experience involving the virtual object that is associated with the recognized to construction model. For example, the process may start a play experience with the identified virtual object, or the process may add the virtual object to an ongoing play experience.

[0360] After completion of the play experience, or responsive to a game event or a user input, the process may return to step S4104 allowing the user to acquire an image of another toy construction model. Alternatively, the process may terminate.

[0361] It will be appreciated that various modifications to the above process may be implemented. For example, the process may recognize parts of a toy construction model and determine whether unlocked virtual objects are associated with one or each of the recognized parts and provide a play experience involving a combination of these unlocked virtual objects, e.g. only if all recognized parts have an unlocked virtual object associated with it. Examples of such a process are described in connection with FIGS. 35-37. The recognized parts may be individual toy construction elements or toy construction models that are interconnected to form a combined model. Alternatively or additionally, the process may restrict use of the unlocked virtual objects, e.g. to a single use or a predetermined number of uses, to certain combinations with other objects, and/or the like.

[0362] Alternatively or additionally, steps S4105 and S4106 may be combined into a single operation. For example, the process may only recognize toy construction models as known toy construction models if they have an unlocked virtual object associated with it.

[0363] Yet alternatively or additionally, in step S4105, the process may further detect an object code applied to a recognized toy construction model, e.g. by reading a QR code or another type of visually recognizable code from the captured image. For example, during manufacturing of a toy construction element, a data processing system executing an encoder may convert a bit string or other object code into a visually recognizable code, such as a QR code, a graphic decoration, and/or the like. The encoded visually recognizable code may then be printed on the toy construction element, e.g. on a torso of a figurine as illustrated in FIGS. 39A-C.

[0364] During step S4105, a decoding function may analyse an image of a toy construction model and extract the object code that was embedded by the encoder. The decoding function may be based on a QR code reading function, a neural network trained to convert encoded images into their object code counterparts, or the like. Error correction codes can be added to the object code so that a number of erroneous output bits can he corrected. In one embodiment, the process may initially recognize the toy construction model, identify a portion of the recognized toy construction model where an object code is expected, and feed a part image depicting the identified portion, e.g. the torso of a recognized figurine to the decoding function.

[0365] In step S4106, in addition to determining the unlocked virtual object corresponding to the recognized toy construction model, the process may further identify a particular instance of the unlocked virtual object based on the detected object code. To this end, the process may maintain records associated with multiple instances of a particular virtual object, each instance being associated with a respective object code and, optionally with respective attributes, such as health, capabilities, etc.

[0366] FIGS. 39A-C illustrate examples of toy construction models 4131. The toy construction models of FIGS. 39A-C are figurines, each constructed from multiple toy construction elements, in particular toy construction elements forming the head, the torso, the legs, respectively, of the figurine. It will be appreciated that, alternatively, each figurine may be formed as a single toy construction element. It will also be appreciated that other toy construction models may represent other items, e.g. a vehicle, a building, an animal, etc. Each figurine has applied to it a computer readable visual code 4735 encoding a serial number or another form of identifier which may uniquely or non-uniquely identify a particular figurine. In the example of FIGS. 39A-C the visual code is printed on the torso of the figurine. However, in other examples the code may be applied to other parts of the model or even be encoded by visual markers applied to respective parts of the model. Hence, even though the figurines 4131 of FIGS. 39A-C have identical shape, size and decoration apart from the code 4735, a computing device having a code reader may distinguish them from each other. Accordingly, as the figurines of FIGS. 39A -C are perceptually very similar to the human observer (in some embodiments they may even be substantially indistinguishable), the end user will not easily notice the difference between two figurines. Moreover, embodiments of the process described herein may recognize the figurines as representing the same virtual object, in particular the same virtual character. However, a single unlock code may unlock all instances of the virtual object.

[0367] FIG. 40 illustrates an example of a use of the toy system described herein, e.g. including figurines as described in connection with FIGS. 39A-C and a suitably programmed portable device 4450, e.g. a tablet or smartphone executing an app that implements a digital game of the toy system. As in the previous examples, the toy construction system includes toy construction elements (not explicitly shown) for constructing a figurine 4131. The toy construction system further comprises a card 4141 including an unlock code 4171.

[0368] Initially, the processing device 4450 reads the unlock code 4171 included, in the toy construction set, e.g. from card 4141. This causes the corresponding virtual object 4451 (in this example a virtual character) to be unlocked. The user may then capture an image of the figurine 4131 carrying one of a set of object codes 4735. The processing device recognises the figurine including the particular code 4735 applied to the figurine. This causes the digital game executed by the processing device 4450 to provide a play experience involving an instance of the virtual character 4451. After completion of the play experience, the user may capture another image of the same or of a different figurine, in particular a figurine resembling figurine 4131 but having a different object code 4735 applied to it. This allows the user to engage in the same or a different play experience involving a different instance of the virtual character. Accordingly, the digital game may store or otherwise maintain game progress (such as health levels, capability levels, or other progress) for respective instances of a virtual character. For example, if two users each have their own figurine with respective object codes, they may both use the processing device 4450 to engage in the digital game using respective instances of the same virtual character, in particular where the virtual character has respective in-game progress.

[0369] FIG. 41 schematically illustrates an example of a toy system described herein. The toy system includes a plurality of toy construction elements 4120 from which one or more toy construction models can be constructed, e.g. as described in connection with FIG. 32. The toy system further comprises two cards 4141, 4142, e.g. made from plastic or cardboard. Each card shows an image or other representation of one of the toy construction models and a machine readable code 4171, 4172, in this example a QR code, which represents an unlock code for a virtual object associated with the respective toy construction model, e.g. a virtual character and a virtual car, respectively. Alternatively, the unlock code(s) may be provided to the user in a different manner, e.g. by mail or sold separately. Hence, each unlock code is a unique code that comes with the product or is given to the user through a physical printed code or a digital code. The unlock code(s) when used (scanned or typed in) then unlocks the possibility of using computer vision to select the object in/for a digital experience.

[0370] The toy system further comprises a suitably programmed processing device 4450, e.g. a tablet or smartphone or other portable computing device executing an app that implements a digital game of the toy system. The processing device 4450 comprises a central processing unit 4455, a memory 4456, a user interface 4457, a code reader 4458 and an image capture device 4459.

[0371] The user interface 4457 may e.g. include a display, such as a touch screen, and, optionally input devices such as buttons, a touch pad, a pointing device, etc.

[0372] The image capture device 4459, may include a digital camera, a depth camera, a stereo camera, and/or the like.

[0373] The code reader 4458 may be a barcode reader, and RFID reader or the like, M some embodiments, the code reader may include a digital camera. In some embodiments, the code reader and the image capture device may be a single device. For example, the same digital camera may be used to read the unlock codes and capture images of the toy construction models.

[0374] FIG. 42 schematically illustrates another example of a toy system described. The toy system of FIG. 42 is similar to the toy system of FIG. 41, the only difference being that the processing device 4450 further comprises a communications interface 4460, such as a wireless or wired communications interface allowing the processing device 4450 to communicate with a remote system 5170. The communication may be wired or wireless. The communication may be via a communication network. The remote system may be a server computer or other suitable data processing system which may be configured to implement one or more of the processing steps described herein. For example, the remote system may maintain a database of unlock codes in order to determine whether a given unlock code has previously been used to unlock a virtual object. Alternatively or additionally, the remote system may maintain a database of object codes. Yet alternatively or additionally, the remote system may implement an object recognition process or parts thereof for recognizing toy construction models in captured images. Yet alternatively or additionally, the remote system may implement at least a part of the digital game, e.g. in embodiments where the digital game includes a multiplayer play experience or a networked play experience.

[0375] Hence, generally, a virtual object needs to be unlocked only once by the unique unlock code. The selection of the virtual object (or of multiple/composite virtual objects at one time) can be done every time the virtual object is to be used in the digital experience or for a one-dine use.

[0376] In the claims enumerating several means, several of these means can be embodied by one and the same element, component or item of hardware. The mere fact that certain measures are recited in mutually different dependent claims or described in different embodiments does not indicate that a combination of these measures cannot be used to advantage.

[0377] It should be emphasized that the term “comprises/comprising” when used in this specification is taken to specify the presence of stated features, elements, steps or components but does not preclude the presence or addition of one or more other features, elements, steps, components or groups thereof.