Anomaly detection method and apparatus, computer device, and computer-readable storage medium
By determining the set of rotation angles of the object to be detected from the dictionary and performing anomaly detection, the problem of inaccurate rotation angle prediction in 3D object detection is solved, the accuracy of the detection results is improved, and precise data support is provided for autonomous driving and augmented reality.
WO2026145370A1PCT designated stage Publication Date: 2026-07-09UBTECH ROBOTICS CORP LTD
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- UBTECH ROBOTICS CORP LTD
- Filing Date
- 2025-12-27
- Publication Date
- 2026-07-09
AI Technical Summary
Technical Problem
Existing technologies for 3D target detection suffer from inaccurate prediction of rotation angles and a lack of anomaly detection methods, resulting in low accuracy of detection results.
Method used
By acquiring the rotation angle in the image to be detected and determining the third rotation angle set of the object to be detected from the M second rotation angle sets in the first dictionary, anomaly detection is performed using the reference rotation angle, thereby improving the accuracy of the rotation angle.
Benefits of technology
It achieves efficient and accurate anomaly detection, providing accurate data support for applications such as autonomous driving, navigation, and augmented reality.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN2025146378_09072026_PF_FP_ABST
Abstract
The present application provides an anomaly detection method and apparatus, a computer device, and a computer-readable storage medium. The method comprises: acquiring an image to be detected, and determining a first identifier of an object to be detected comprised in said image and a first rotation angle of a bounding box of said object; acquiring a first dictionary, wherein the first dictionary comprises M second identifiers and a second rotation angle set corresponding to each second identifier, and the second rotation angle set comprises second rotation angles corresponding to different timestamps; determining a third rotation angle set corresponding to the first identifier from among the M second rotation angle sets; when the quantity of elements in the third rotation angle set is greater than a first quantity threshold, determining, from the third rotation angle set, a reference rotation angle corresponding to the first identifier; and performing anomaly detection on the first rotation angle on the basis of the reference rotation angle to obtain a detection result of the first rotation angle. By means of the present application, the anomaly detection of rotation angles can be achieved.
Need to check novelty before this filing date? Find Prior Art