Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) |
| Herausgeber (Verlag) | Institute of Electrical and Electronics Engineers Inc. |
| Seiten | 4191-4201 |
| Seitenumfang | 11 |
| ISBN (elektronisch) | 9798350307443 |
| ISBN (Print) | 9798350307450 |
| Publikationsstatus | Veröffentlicht - 2023 |
| Veranstaltung | 2023 IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2023 - Paris, Frankreich Dauer: 2 Okt. 2023 → 6 Okt. 2023 |
Abstract
Recently, regression-based methods have dominated the field of 3D human pose and shape estimation. Despite their promising results, a common issue is the misalignment between predictions and image observations, often caused by minor joint rotation errors that accumulate along the kinematic chain. To address this issue, we propose to construct dense correspondences between initial human model estimates and the corresponding images that can be used to refine the initial predictions. To this end, we utilize renderings of the 3D models to predict per-pixel 2D displacements between the synthetic renderings and the RGB images. This allows us to effectively integrate and exploit appearance information of the persons. Our per-pixel displacements can be efficiently transformed to per-visible-vertex displacements and then used for 3D model refinement by minimizing a reprojection loss. To demonstrate the effectiveness of our approach, we refine the initial 3D human mesh predictions of multiple models using different refinement procedures on 3DPW and RICH. We show that our approach not only consistently leads to better image-model alignment, but also to improved 3D accuracy.
ASJC Scopus Sachgebiete
- Informatik (insg.)
- Artificial intelligence
- Informatik (insg.)
- Angewandte Informatik
- Informatik (insg.)
- Maschinelles Sehen und Mustererkennung
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW). Institute of Electrical and Electronics Engineers Inc., 2023. S. 4191-4201.
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Personalized 3D Human Pose and Shape Refinement
AU - Wehrbein, Tom
AU - Rosenhahn, Bodo
AU - Matthews, Iain
AU - Stoll, Carsten
N1 - Funding Information: This work was supported by the Federal Ministry of Education and Research (BMBF), Germany, under the project LeibnizKILabor (grant no. 01DD20003) and the AI service center KISSKI (grant no. 01IS22093C), the Center for Digital Innovations (ZDIN) and the Deutsche Forschungsgemeinschaft (DFG) under Germany's Excellence Strategy within the Cluster of Excellence PhoenixD (EXC 2122).
PY - 2023
Y1 - 2023
N2 - Recently, regression-based methods have dominated the field of 3D human pose and shape estimation. Despite their promising results, a common issue is the misalignment between predictions and image observations, often caused by minor joint rotation errors that accumulate along the kinematic chain. To address this issue, we propose to construct dense correspondences between initial human model estimates and the corresponding images that can be used to refine the initial predictions. To this end, we utilize renderings of the 3D models to predict per-pixel 2D displacements between the synthetic renderings and the RGB images. This allows us to effectively integrate and exploit appearance information of the persons. Our per-pixel displacements can be efficiently transformed to per-visible-vertex displacements and then used for 3D model refinement by minimizing a reprojection loss. To demonstrate the effectiveness of our approach, we refine the initial 3D human mesh predictions of multiple models using different refinement procedures on 3DPW and RICH. We show that our approach not only consistently leads to better image-model alignment, but also to improved 3D accuracy.
AB - Recently, regression-based methods have dominated the field of 3D human pose and shape estimation. Despite their promising results, a common issue is the misalignment between predictions and image observations, often caused by minor joint rotation errors that accumulate along the kinematic chain. To address this issue, we propose to construct dense correspondences between initial human model estimates and the corresponding images that can be used to refine the initial predictions. To this end, we utilize renderings of the 3D models to predict per-pixel 2D displacements between the synthetic renderings and the RGB images. This allows us to effectively integrate and exploit appearance information of the persons. Our per-pixel displacements can be efficiently transformed to per-visible-vertex displacements and then used for 3D model refinement by minimizing a reprojection loss. To demonstrate the effectiveness of our approach, we refine the initial 3D human mesh predictions of multiple models using different refinement procedures on 3DPW and RICH. We show that our approach not only consistently leads to better image-model alignment, but also to improved 3D accuracy.
UR - http://www.scopus.com/inward/record.url?scp=85182935057&partnerID=8YFLogxK
U2 - 10.1109/ICCVW60793.2023.00453
DO - 10.1109/ICCVW60793.2023.00453
M3 - Conference contribution
AN - SCOPUS:85182935057
SN - 9798350307450
SP - 4191
EP - 4201
BT - 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2023
Y2 - 2 October 2023 through 6 October 2023
ER -