The present disclosure relates to an image processing apparatus and method for transforming image data of a spherical coordinate system corresponding to image data of a planar coordinate system without transforming the image data of the planar coordinate system, and stitching the transformed image data of the spherical coordinate system to the image data of the planar coordinate system.

Presented are systems and methods for 3D reconstruction from videos using an unsupervised learning framework for depth and normal estimation via edge-aware depth-normal consistency. In embodiments, this is accomplished by using a surface normal representation. Depths may be reconstructed in a single image by watching unlabeled videos. Depth-normal regularization constrains estimated depths to be compatible with predicted normals, thereby, yielding geometry-consistency and improving evaluation performance and training speed. In embodiments, a consistency term is solved by constructing depth-to-normal layer and normal-to-depth layers within a deep convolutional network (DCN). In embodiments, the depth-to-normal layer uses estimated depths to compute normal directions based on neighboring pixels. Given the estimated normals, the normal-to-depth layer may then output a regularized depth map. Both layers may be computed with awareness of edges within the image. Finally, to train the network, the photometric error and gradient smoothness for both depth and normal predictions may be applied.

A detection system for a vehicle is provided. The detection system may include at least one image capture device configured to acquire a plurality of images of an area forward of the vehicle, the area including a curb separating a road surface from an off-road surface and a data interface. The detection system may also include at least one processing device programmed to receive the plurality of images via the data interface, and determine a plurality of curb edge line candidates in the plurality of images. The at least one processing device may be further programmed to identify at least one edge line candidate as an edge line of the curb.

Disclosed is a method and apparatus for calibrating parameters of a three-dimensional (3D) display apparatus, the method including acquiring a first captured image of a 3D display apparatus displaying a first pattern image, adjusting a first parameter set of the 3D display apparatus based on the first captured image, acquiring a second captured image of the 3D display apparatus displaying a second pattern image based on the adjusted first parameter set, and adjusting a second parameter set of the 3D display apparatus based on the second captured image.

Systems and methods are disclosed method for rendering reality for an object by performing motion tracking and area learning of an environment; selecting a pattern or color from a plurality of product variations; blending the pattern or color of the object and the environment; and displaying the color on the object as an augmented reality view.

According to an example aspect of the present invention, there is provided for an apparatus comprising a User Interface, UI, the method comprising generating or receiving particles, wherein each particle has a first position associated with a first shape, and storing the first positions as current positions, determining a second shape, wherein the second shape forms a first target shape for a first transition from the first shape, and a first target time for the first transition, generating and storing first target positions of the particles associated with the second shape, comparing the first target time to a current time, responsive to the first target time being larger than the current time, determining intermediate positions of the particles, based on the current positions and the first target positions of the particles, wherein the intermediate positions are closer to the target positions than the current positions and updating the user interface by using the intermediate positions of the particles to provide a graphical output on a display, storing the intermediate positions as the current positions.

A video management system may identify moving objects in a scene, and may obscure the moving object. Additionally, the video management system may identify parts of the scene that are not moving, and may identify pixels having a color falling within a range which may be associated with human skin, and may obscure such pixels. The video management system may present an image to a monitor including the obscured moving object and the obscured pixels having the color falling within the range associated with human skin.

Systems and methods for generating high resolution dewarped images for an image of a document captured by a 3D stereo digital camera system, or a mobile phone camera capturing a sequence of images, which may improve OCR performance. Example embodiments include a compact stereo camera with two sensors mounted at fixed locations, and a multi-resolution pipeline to process and to dewarp the images using a three dimensional surface model based on curve profiles of the computed depth map. Example embodiments also include a mobile phone including a camera which captures a sequence of images, and a processor which computes a disparity map using the captured sequence of image frames, computes a model of the at least one document page by generating a cylindrical three dimensional geometric surface using the computed disparity map, and renders a dewarped image from the computed model.

A method and system provide the ability to modify a three-dimensional (3D) model. The 3D model is obtained and arbitrary faces of the model are selected. Shape modification operations to be performed are prescribed. A deformation lattice is constructed by setting up a lattice structure with control points. A space of the 3D model is mapped to a space of the lattice structure. The deformation lattice is evaluated by deforming the lattice using a selected set of control points. The evaluated deformed model is then output.

An electronic device, method, and computer readable medium for foveated storage and processing are provided. The electronic device includes a memory, and a processor coupled to the memory. The processor performs head tracking and eye tracking; generates a foveated image from an original image based on the head tracking and the eye tracking; and stores the foveated image using one of: a tile-based method or a frame-based method.