Pedestrian pose resolving method for vehicle

Abstract

The invention discloses a pedestrian pose resolving method for a vehicle. The method comprises the following steps: a plurality of vehicles which can be networked performing real-time perception on pedestrians in a visual field range by using all-round cameras to obtain pedestrian three-dimensional point clouds of a single-vehicle visual field; filtering invalid vehicles according to the driving speed return value of the vehicle-mounted system to form a local area Internet of Vehicles capable of sharing valid data; synthesizing and splicing pedestrian three-dimensional point clouds of multiplevehicle views based on the local area Internet of Vehicles to obtain global three-dimensional point clouds; obtaining a global pedestrian three-dimensional pose by utilizing a position relationship among all points in the point cloud; sharing the global pedestrian three-dimensional pose to all vehicles in the local Internet of Vehicles; and based on the bicycle safety range, analyzing the globalpedestrian three-dimensional pose to obtain pedestrian early warning information in the bicycle safety range. By utilizing the method, the perception range of the vehicle can be expanded in scenes such as parking lots and intersections, and meanwhile, the vehicle without perception capability can perceive pedestrian information in a safety range.

Description

technical field [0001] The invention relates to a pose judgment method, in particular to a pedestrian pose solution method for vehicles. Background technique [0002] At present, most automobiles and mobile robots already have the ability to detect pedestrians based on vision, but the detection results are two-dimensional bounding boxes. Two-dimensional data cannot provide the spatial position information of the object, and the detection result lacks a three-dimensional sense of space, and cannot accurately perceive the real three-dimensional pose of the object in space. At present, the detection of the three-dimensional pose of an object is mainly completed by lidar, binocular camera, ToF, and optically encoded camera. The above hardware has certain requirements on the light intensity in the environment and the imaging distance of the object. In a real driving scene, there will inevitably be blind spots in the field of vision due to vehicle occlusion and building occlusion...

AI Technical Summary

Problems solved by technology

[0002] At present, most cars and mobile robots have the ability to detect pedestrians based on vision, but the detection results are two-dimensional bounding boxes

The two-dimensional data cannot provide the spatial position information of the object, the detection result lacks a three-dimensional sense of space, and cannot accurately perceive the real three-dimensional pose of the object in space

Traditional bicycle perception cannot perceive pedestrian information in blind spots

In traditional multi-vehicle collaborative tasks, 3D point cloud merging is a method with little meaning. The main reason is that the accuracy of the point cloud is too low to extract the key points of the object based on the point cloud, and even the type of the object cannot be judged.

[0003] Therefore, the existing pedestrian detection technology has problems such as inability to perceive the real 3D position of pedestrians, low imaging quality, limited perception range, and low point cloud accuracy.

Images