A method, apparatus, and system for estimating a moving speed of a detected pedestrian at an autonomous driving vehicle (ADV) is disclosed. A pedestrian is detected in a plurality of frames of point clouds generated by a LIDAR device installed at an autonomous driving vehicle (ADV). In each of at least two of the plurality of frames of point clouds, a minimum bounding box enclosing points corresponding to the pedestrian excluding points corresponding to limbs of the pedestrian is generated. A moving speed of the pedestrian is estimated based at least in part on the minimum bounding boxes across the at least two of the plurality of frames of point clouds. A trajectory for the ADV is planned based at least on the moving speed of the pedestrian. Thereafter, control signals are generated to drive the ADV based on the planned trajectory.

Search points in a search space may be projected onto images from cameras imaging different parts of the search space. Subimages, corresponding to the projected search points, may be selected and processed to determine if a target object has been detected. Based on subimages in which target objects are detected, as well as orientation data from cameras capturing images from which the subimages were selected, positions of the target objects in the search space may be determined.

A vehicle navigation image synthesis method includes: acquiring M sampling images; storing N sampling images, a value of N being determined according to a speed of a target vehicle, a position of the target vehicle, and a following distance of the target vehicle; determining a plurality of sampling points of each sampling image of the N sampling images; determining first coordinates of each sampling point of the plurality of sampling points of each sampling image; mapping the first coordinates of each to the reference coordinate system to determine a second coordinate of each sampling point; and synthesizing sampling points of the N sampling images according to second coordinates of the sampling points of the N sampling images to determine a target image.

Provided is an information processing apparatus including an image supply unit that supplies a plurality of input images showing corresponding objects to an image processing unit and obtains a plurality of object images as an image processed result from the image processing unit, and a display control unit that synchronously displays the plurality of object images that have been obtained. The object images are regions including the corresponding objects extracted from the plurality of input images, and orientations, positions, and sizes of the corresponding objects of the plurality of object images are unified.

A sensor system that obtains and processes time-of-flight data (TOF) obtained in an arbitrary orientation is provided. A TOF sensor obtains distance data describing various surfaces. A processor identifies a horizontal Z-plane in the environment, and transforms the data to align with the Z-plane. In some embodiments, the environment includes a box, and the processor identifies a bottom and a top of the box in the transformed data. The processor can further determine dimensions of the box, e.g., the height between the top and bottom of the box, and the length and width of the box top.

A system and method for estimating dimensions of an approximately cuboidal object from a 3D image of the object acquired by an image sensor of the vision system processor is provided. An identification module, associated with the vision system processor, automatically identifies a 3D region in the 3D image that contains the cuboidal object. A selection module, associated with the vision system processor, automatically selects 3D image data from the 3D image that corresponds to approximate faces or boundaries of the cuboidal object. An analysis module statistically analyzes, and generates statistics for, the selected 3D image data that correspond to approximate cuboidal object faces or boundaries. A refinement module chooses statistics that correspond to improved cuboidal dimensions from among cuboidal object length, width and height. The improved cuboidal dimensions are provided as dimensions for the object. A user interface displays a plurality of interface screens for setup and runtime operation.

A method includes creating a point cloud model of an environment, applying at least one filter to the point cloud model to produce a filtered model of the environment and defining a plane in the filtered model corresponding to a horizontal expanse associated with a floor of the environment.

Search points in a search space may be projected onto images from cameras imaging different parts of the search space. Subimages, corresponding to the projected search points, may be selected and processed to determine if a target object has been detected. Based on subimages in which target objects are detected, as well as orientation data from cameras capturing images from which the subimages were selected, positions of the target objects in the search space may be determined.

This application discloses an image processing method performed by a computer device. In this embodiment of this application, feature point recognition can be performed on a face image to obtain a plurality of facial feature points of the face image; feature point position offset information between the feature points and reference facial feature points of a reference face image is determined; based on the feature point position offset information, position adjustment is performed on a facial feature point of a reference face depth image corresponding to the reference face image to obtain a target face depth image corresponding to the face image; and direction deflection is performed on the face image according to the target face depth image to obtain a target face image.

A method for determining a collision range includes: determining a grid plane corresponding to a model, the grid plane including a plurality of grid regions; projecting the model to the grid plane and obtaining height information of the model with respect to the grid plane, the height information of the model including grid height information corresponding to a grid region that identifies a height of a sub-model in the model with respect to the grid plane, and the sub-model being a portion of the model in a projection space of the grid region; and obtaining a collision range of the model according to the grid plane and the height information.