It is possible to determine a geometric transformation matrix representing geometric transformation between an input image and a template image with high precision. A geometric transformation matrix/inlier estimation section 32 determines a corresponding point group serving as inliers, and estimates the geometric transformation matrix representing the geometric transformation between the input image and the template image. A scatter degree estimation section 34 estimates scatter degree of the corresponding points based on the corresponding point group serving as inliers. A plane tracking convergence determination threshold calculation section 36 calculates a threshold used in convergence determination when iterative update of the geometric transformation matrix in a plane tracking section 38 is performed based on the estimated scatter degree. The plane tracking section 38 iterates update of the geometric transformation matrix so as to minimize a difference in value to pixels of a geometric transformation image obtained by transforming one of the input image and the templated image using the geometric transformation matrix and corresponding pixels until it is determined that convergence has been completed using the calculated threshold.

A system and method for human motion detection and tracking are disclosed. In one embodiment, a smart device having an optical sensing instrument monitors a stage. Memory is accessible to a processor and communicatively coupled to the optical sensing instrument. The system captures an image frame from the optical sensing instrument. The image frame is then converted into a designated image frame format, which is provided to a pose estimator. A two-dimensional dataset is received from the pose estimator. The system then converts, using inverse kinematics, the two-dimensional dataset into a three-dimensional dataset, which includes time-independent static joint positions, and then calculates, using the three-dimensional dataset, the position of each of the respective plurality of body parts in the image frame.

A system and method for human motion detection and tracking are disclosed. In one embodiment, a smart device having an optical sensing instrument monitors a stage. Memory is accessible to a processor and communicatively coupled to the optical sensing instrument. The system captures an image frame from the optical sensing instrument. The image frame is then converted into a designated image frame format, which is provided to a pose estimator. A two-dimensional dataset is received from the pose estimator. The system then converts, using inverse kinematics, the two-dimensional dataset into a three-dimensional dataset, which includes time-independent static joint positions, and then calculates, using the three-dimensional dataset, the position of each of the respective plurality of body parts in the image frame.

The technical solutions described herein pertain to apparatus and methods for accurately determining in cricket whether a ball is above the waist, shoulder, or head of a batter standing upright at the popping crease. Thus, certain embodiments of the technical solutions described herein provide apparatus and methods for accurately determining in cricket whether a ball is a no-ball, a wide-ball, or a legal ball. Certain embodiments of the technical solutions described herein also pertain to determining whether a legal ball is a bouncer. Particularly, the technical solutions described herein provide apparatus and methods for objectively determining in cricket whether a ball is a no-ball, a wide-ball, or a legal ball.

The technical solutions described herein pertain to apparatus and methods for accurately determining in cricket whether a ball is above the waist, shoulder, or head of a batter standing upright at the popping crease. Thus, certain embodiments of the technical solutions described herein provide apparatus and methods for accurately determining in cricket whether a ball is a no-ball, a wide-ball, or a legal ball. Certain embodiments of the technical solutions described herein also pertain to determining whether a legal ball is a bouncer. Particularly, the technical solutions described herein provide apparatus and methods for objectively determining in cricket whether a ball is a no-ball, a wide-ball, or a legal ball.

The invention is related to a method of estimating an orientation of an object in an image, comprising the steps of: calculating, for the object in the image, a probability distribution of rotation; and estimating the orientation of the object from the calculated probability distribution; wherein the step of calculating the probability distribution and/or the step of estimating the orientation of the object are executed by a neural network; wherein the probability distribution is a matrix Fisher probability density function; and wherein the step of calculating the probability distribution includes approximating a normalizing function for the matrix Fisher probability density function.

This application relates to a target object detection method and apparatus, a non-transitory computer-readable storage medium, and a computer device. The method includes: obtaining a to-be-detected image; inputting the to-be-detected image into a target object detection model; generating, by the target object detection model, a prediction diagram corresponding to the to-be-detected image, the prediction diagram describing a relation degree to which pixels of the to-be-detected image belong to a target detection object; and performing region segmentation on the prediction diagram to obtain a target detection object region. In addition, a method and an apparatus for training an object detection model into the target object detection model, a non-transitory computer-readable storage medium, and a computer device are also provided.

This application relates to a target object detection method and apparatus, a non-transitory computer-readable storage medium, and a computer device. The method includes: obtaining a to-be-detected image; inputting the to-be-detected image into a target object detection model; generating, by the target object detection model, a prediction diagram corresponding to the to-be-detected image, the prediction diagram describing a relation degree to which pixels of the to-be-detected image belong to a target detection object; and performing region segmentation on the prediction diagram to obtain a target detection object region. In addition, a method and an apparatus for training an object detection model into the target object detection model, a non-transitory computer-readable storage medium, and a computer device are also provided.

The invention relates to a method for improving localization accuracy of a self-driving vehicle (100). The method comprises steps of receiving from one or more range sensing devices (110) point cloud data related to surface (130) characteristics of an environment of a self-driving vehicle (100), and based on receiving, constructing a modified normal distributions transform (NDT) histogram having a set of Gaussian distributions in a plurality of histogram bins, each of the plurality of histogram bins providing different constraining features, performing subsampling for each histogram bins in the constructed NDT histogram, in which subsampling a number of Gaussian distributions from each histogram bin is removed to construct a vector h.sup.S representing the target height of each histogram bin, and after subsampling, selecting h.sub.i.sup.S Gaussian distributions from the corresponding histogram bins of vector h.sup.S based on the constraining features given by the Gaussian distributions and adding them to the subsample set S in order to localize the self-driving vehicle (100)) with respect to the point cloud data received.

Techniques for resolving a range to an object using histograms are disclosed. A frame collection time for a depth-image frame is divided into a plurality of different collection subframes, where each collection subframe encompasses a plurality of light pulse cycles. Counts of accumulated photon detections by a pixel during the different collection subframes are allocated to histogram bins using different bin maps for the collection subframes. Each bin map defines a different mapping of time to bins for the light pulse cycles within its applicable collection subframe, and each mapping defines a bin width for its bins so that its bin map covers a maximum detection range for the depth-image frame. A range to an object in the pixel's field of view (within the maximum detection range) can be resolved according to a combination of peak bin positions in the histogram data with respect to the different collection subframes.