Methods and devices for determining gaze area; vehicle control methods and devices

By simultaneously acquiring facial images using a multi-view image acquisition device, and combining 3D facial key points with gaze information, the problem of insufficient accuracy in gaze region detection is solved, achieving higher accuracy in gaze region determination.

CN115035499BActive Publication Date: 2026-05-26SHANGHAI SENSETIME LINGANG INTELLIGENT TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHANGHAI SENSETIME LINGANG INTELLIGENT TECH CO LTD
Filing Date
2022-05-31
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately determine the gaze area, especially in autonomous driving and smart home scenarios where external factors have a significant impact, leading to insufficient accuracy in gaze area information.

Method used

Multiple facial images are simultaneously acquired using image acquisition devices from different perspectives. By combining the three-dimensional position information of facial key points with the fused gaze information, the gaze area is determined.

Benefits of technology

It improves the accuracy of fixation region information, reduces the influence of external factors on fixation region determination, and enhances the accuracy of fixation region detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115035499B_ABST
    Figure CN115035499B_ABST
Patent Text Reader

Abstract

This disclosure provides a method, vehicle control method, and apparatus for determining a gaze region. The method for determining the gaze region includes: acquiring multiple facial images of a target object simultaneously acquired by multiple image acquisition devices from different perspectives within a predetermined space; determining three-dimensional position information of predetermined facial key points of the target object based on at least two of the facial images; determining fused gaze information of the target object based on the multiple facial images; and determining gaze region information of the target object within the predetermined space based on the three-dimensional position information of the predetermined facial key points and the fused gaze information.
Need to check novelty before this filing date? Find Prior Art