Pose information determination method, apparatus, device, and storage medium

By using feature element combination and matrix calculation in vehicle pose information determination, the shortcomings of brightness feature points and optical flow tracing in the prior art are solved, and higher accuracy and stability of pose information are achieved.

CN117095377BActive Publication Date: 2026-07-21TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TENCENT TECHNOLOGY (SHENZHEN) CO LTD
Filing Date
2022-05-10
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing technologies, when determining vehicle pose information, suffer from low matching degree and poor accuracy because feature points with high brightness lack semantic information and optical flow tracing is greatly affected by illumination.

Method used

By determining the combination of feature elements in the first road image and multiple second road images of the target vehicle, matching is performed using feature points of fixed image elements, and pose offset information is calculated by combining homography matrix and fundamental matrix, thereby determining the actual pose of the vehicle.

Benefits of technology

It improves the accuracy and applicability of vehicle pose information, reduces the impact of lighting and environmental changes, and enhances the stability and accuracy of feature point matching.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117095377B_ABST
    Figure CN117095377B_ABST
Patent Text Reader

Abstract

Embodiments of the present application disclose a pose information determination method and device, equipment and a storage medium, which can be applied to the fields of map, navigation, automatic driving, Internet of Vehicles, intelligent transportation, artificial intelligence and cloud computing. The method comprises the following steps: determining a first road image corresponding to a target vehicle at a first time and at least one second road image corresponding to the target vehicle at a past second time; determining each first feature element of the first road image and each second feature element in each second road image; determining a plurality of element combinations based on the first feature elements and the second feature elements; determining pose bias information of the target vehicle at the first time based on the element combinations; and determining actual pose information of the target vehicle at a next time based on actual pose information of the target vehicle at the first time and the pose bias information. According to the embodiments of the present application, the actual pose information of the target vehicle can be accurately determined, and the applicability is high.
Need to check novelty before this filing date? Find Prior Art