Perspective Crop Based Egocentric Hand Pose Estimation via Fisheye Stereo Vision
| dc.contributor.author | Hur, Hyejin | en_US |
| dc.contributor.author | Baek, Seongmin | en_US |
| dc.contributor.author | Gil, Younhee | en_US |
| dc.contributor.author | Kim, Sangpil | en_US |
| dc.contributor.editor | Günther, Tobias | en_US |
| dc.contributor.editor | Montazeri, Zahra | en_US |
| dc.date.accessioned | 2025-05-09T09:31:03Z | |
| dc.date.available | 2025-05-09T09:31:03Z | |
| dc.date.issued | 2025 | |
| dc.description.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. | en_US |
| dc.description.sectionheaders | Posters | |
| dc.description.seriesinformation | Eurographics 2025 - Posters | |
| dc.identifier.doi | 10.2312/egp.20251016 | |
| dc.identifier.isbn | 978-3-03868-269-1 | |
| dc.identifier.issn | 1017-4656 | |
| dc.identifier.pages | 2 pages | |
| dc.identifier.uri | https://doi.org/10.2312/egp.20251016 | |
| dc.identifier.uri | https://diglib.eg.org/handle/10.2312/egp20251016 | |
| dc.publisher | The Eurographics Association | en_US |
| dc.rights | Attribution 4.0 International License | |
| dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
| dc.subject | CCS Concepts: Computing methodologies → Computer vision; Vision for robotics | |
| dc.subject | Computing methodologies → Computer vision | |
| dc.subject | Vision for robotics | |
| dc.title | Perspective Crop Based Egocentric Hand Pose Estimation via Fisheye Stereo Vision | en_US |