The present disclosure relates to a computer-implemented method for determining a central point of point cloud data in an automotive system for monitoring the environment of a vehicle. The point cloud data is generated by one or more sensors of the vehicle with respect to a reference coordinate system. The point cloud data defines a connected subspace of the reference coordinate system. The method comprises a step a) of determining a bounding box of the point cloud data, a step b) of selecting a starting agent position of an agent within the bounding box, and a step c) of selecting a coordinate system relative to the bounding box. In a further step d) a plurality of agent moving operations are performed. Each agent moving operation comprises moving the agent from the current agent position to a new agent position parallel to a coordinate axis of the selected coordinate system. The new agent position is determined based on an intersecting line through the current agent position parallel to the coordinate axis. The method further comprises a step e) of determining, after step d) is completed, the new agent position as the central point of the point cloud data.

A computer-implemented system and method includes generating first pseudo segment data from a first augmented image and generating second pseudo segment data from a second augmented image. The first augmented image and the second augmented image are in a dataset along with other augmented images. A machine learning system is configured to generate pixel embeddings based on the dataset. The first pseudo segment data and the second pseudo segment data are used to identify a first set of segments that a given pixel belongs with respect to the first augmented image and the second augmented image. A second set of segments is identified across the dataset. The second set of segments do not include the given pixel. A local segmentation loss is computed for the given pixel based on the corresponding pixel embedding that involves attracting the first set of segments while repelling the second set of segments. The global concept loss is computed based on a similarity determination between the first set of segments and a concept vector of a corresponding concept. The corresponding concept categorizes the first set of segments with other sets of segments across the dataset based on semantic meaning. The parameters of machine learning system are updated based on the total loss that takes into account at least the local segmentation loss and the global concept loss.

According to one example for outputting image data, an image comprising a surface and an object are captured on a sensor. An object mask based on the captured image is created on a processor. A first composite image based on the object mask and a source content file is created. In an example, the first composite image is projected to the surface.

An image evaluation method and an electronic apparatus thereof are provided. The method is adapted to the electronic apparatus installed at a vehicle for evaluating whether the vehicle is deviated from a lane, and the electronic apparatus includes at least one image capture unit. The method includes the following steps. An image related to driving environment is obtained by the image capture unit. According to an evaluation criterion, a plurality of candidate pair and corresponding candidate points in the image are evaluated from an arbitrary row of pixels of the image to recognize lane stripes in the image. A corresponding position of a feature portion of the vehicle corresponding to the image is calculated. The corresponding position of the feature portion corresponding to the image is compared with the lane stripes to determine whether the vehicle is towards or crosses over real lane stripes.

Human Computer Interfaces (HCl) may allow a user to interact with a computer via a variety of mechanisms, such as hand, head, and body gestures. Various of the disclosed embodiments allow information captured from a depth camera on an HCl system to be used to recognize such gestures. Particularly, the HCl system's depth sensor may capture depth frames of the user's movements over time. To discern gestures from these movements, the system may group portions of the user's anatomy represented by the depth data into classes. “Features” which reflect distinguishing features of the user's anatomy may be used to accomplish this classification. Some embodiments provide improved systems and methods for generating and/or selecting these features. Features prepared by various of the disclosed embodiments may be less susceptible to overfitting training data and may more quickly distinguish portions of the user's anatomy.

An apparatus includes a display control unit, a receiving unit, an adjusting unit, and a determination unit. The display control unit is configured to display an image showing a result of detection of a defect from a captured image of a structure on a display device. The receiving unit is configured to receive an operation to specify part of the displayed image as a first region and an operation to give an instruction to correct at least part of the detection data corresponding to the first region. The adjusting unit is configured to adjust a parameter to be applied to the first region according to the instruction. The determination unit is configured to determine one or more second regions to which the adjusted parameter is to be applied from a plurality of segmented regions of the image.

An image matching system for determining visual overlaps between images by using box embeddings is described herein. The system receives two images depicting a 3D surface with different camera poses. The system inputs the images (or a crop of each image) into a machine learning model that outputs a box encoding for the first image and a box encoding for the second image. A box encoding includes parameters defining a box in an embedding space. Then the system determines an asymmetric overlap factor that measures asymmetric surface overlaps between the first image and the second image based on the box encodings. The asymmetric overlap factor includes an enclosure factor indicating how much surface from the first image is visible in the second image and a concentration factor indicating how much surface from the second image is visible in the first image.

A computing device of a first vehicle may receive a first image and a second image of a second vehicle having flashing light signals. The computing device may determine, in the first image and the second image, an image region that bounds the second vehicle such that the image region substantially encompasses the second vehicle. The computing device may determine a polar grid that partitions the image region in the first image and the second image into polar bins, and identify portions of image data exhibiting a change in color and a change in brightness between the first image and the second image. The computing device may determine a type of the flashing light signals and a type of the second vehicle; and accordingly provide instructions to control the first vehicle.

A head mounted display (HMD) comprises an eye tracking system configured to perform a calibration process using an eye tracking system of the HMD that includes determining a pupillary axis and/or determining an angular offset between the pupillary axis and the eye's true line of sight. The eye tracking system obtains an eye model captures images of the user's pupil while the user is looking at a target or other content displayed on the HMD. In some embodiments, the calibration process is based on a single image of the user's eye and is performed only once. For example, the process can be performed the first time the user uses the HMD, which stores the calibration data for the user in a memory for future use.

A harmonic densely connecting method includes an input step, a plurality of layer operation steps and an output step. The input step is for storing an original input tensor of the block into a memory. Each of the layer operation steps includes a layer-input tensor concatenating step and a convolution operation step. The layer-input tensor concatenating step is for selecting at least one layer-input element tensor of a layer-input set from the memory according to an input connection rule. When a number of the at least one layer-input element tensor is greater than 1, concatenating all of the layer-input element tensors and producing a layer-input tensor. The convolution operation step is for calculating a convolution operation to produce at least one result tensor and then storing the at least one result tensor into the memory. The output step is for outputting a block output.