Perspective Crop Based Egocentric Hand Pose Estimation via Fisheye Stereo Vision

Loading...
Thumbnail Image
Date
2025
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
In this paper, we propose a method to improve the performance of hand pose estimation from egocentric view. To accurately capture hands moving within a wide range in daily activities, we mounted a fisheye stereo camera on a head mounted display to obtain wide-angle images from egocentric view. Our proposed two-stage method addresses the camera distortion introduced by this setup. The 2D hand keypoints estimated by stage-1 HandNet are converted into 3D hand keypoints through triangulation for perspective cropping. Stage-2 HandNet then predicts the final 2D hand keypoints from the undistorted hand crop image. To train stage-1 HandNet for perspective cropping, we built FisheyeEgoHAND dataset which consists of three categories of scenarios (separate hand, hand-hand, and hand-object) that reflect various hand interactions in an egocentric view. Through experiments, we demonstrated that two-stage 2D hand pose estimation outperforms one-stage approach without perspective cropping.
Description

CCS Concepts: Computing methodologies → Computer vision; Vision for robotics

        
@inproceedings{
10.2312:egp.20251016
, booktitle = {
Eurographics 2025 - Posters
}, editor = {
Günther, Tobias
and
Montazeri, Zahra
}, title = {{
Perspective Crop Based Egocentric Hand Pose Estimation via Fisheye Stereo Vision
}}, author = {
Hur, Hyejin
and
Baek, Seongmin
and
Gil, Younhee
and
Kim, Sangpil
}, year = {
2025
}, publisher = {
The Eurographics Association
}, ISSN = {
1017-4656
}, ISBN = {
978-3-03868-269-1
}, DOI = {
10.2312/egp.20251016
} }
Citation