A monocular three-dimensional object detection method, device, equipment and product

By acquiring two-dimensional center point and depth information through the CenterNet network model, three-dimensional center point and orientation angle are generated, which solves the problem of difficult detection of truncated objects in monocular 3D detection and realizes accurate detection of truncated objects and stable perception of close-range targets of vehicles.

CN113887289BActive Publication Date: 2025-11-07JILUO TECH (SHANGHAI) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111013230.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-08-31
Publication Date
2025-11-07
Estimated Expiration
2041-08-31

AI Technical Summary

Technical Problem

Existing monocular 3D detection technology exhibits significant differences in visibility when detecting truncated objects, making it difficult to accurately obtain their 3D center point, resulting in poor detection performance.

Method used

By employing the CenterNet network model, the three-dimensional center point and orientation angle of the object to be detected are generated by acquiring the two-dimensional center point and depth information of the object, thereby achieving accurate detection of truncated objects.

Benefits of technology

It significantly optimizes the prediction performance of truncated objects in monocular 3D detection, improves the accuracy of vehicle near-range target perception, and can perform stable 3D detection of both normal and truncated objects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113887289B_ABST
    Figure CN113887289B_ABST
Patent Text Reader

Abstract

The application provides a monocular three-dimensional object detection method, device, equipment and product, and relates to the technical field of computer vision.The method comprises the following steps: acquiring a to-be-detected image; inputting the to-be-detected image into a monocular three-dimensional object detection model to obtain a two-dimensional detection result, object depth, observation angle and size of the to-be-detected object output by the monocular three-dimensional object detection model; generating a three-dimensional center point of the to-be-detected object in a three-dimensional space according to a two-dimensional center point and the object depth; generating an orientation angle of the to-be-detected object according to the three-dimensional center point and the observation angle; and generating a three-dimensional detection result according to the three-dimensional center point, the orientation angle and the size.The application significantly optimizes the prediction performance of a truncated object in monocular three-dimensional detection, is not limited to the type of a detection object, and can perform monocular three-dimensional detection on normal objects and truncated objects.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Three-dimensional target detection system and method based on binocular vision

    CN111563415A

  • Monocular three-dimensional object detection method and device based on decoupling truncation object

    CN112712062A