Research map / 3D vision / Human pose estimation
Learning To Regress Bodies From Images Using Differentiable Semantic Rendering
Sai Kumar Dwivedi, Nikos Athanasiou, Muhammed Kocabas, Michael J. Black
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add DSR-C and DSR-MC losses (ablation) | C-EFT regressor · 3DPW | PA-MPJPE: 58.5 -> 57.2 | Improved |
| add DSR-C and DSR-MC losses (ablation) | C-EFT regressor · 3DPW | PVE: 119.3 -> 116.3 | Improved |
| add full-body silhouette per-pixel supervision (DSR-FB) (ablation) | C-EFT · 3DPW | PA-MPJPE: 58.5 -> 59.8 | Hurt |
| use soft-DistM loss for full-body silhouette (DSR-FB s-DistM) (ablation) | C-EFT · 3DPW | PA-MPJPE: 58.5 -> 58.0 | Improved |
| add clothing semantics (DSR-C + DSR-MC) over cloth-agnostic DSR-FB s-DistM (ablation) | C-EFT · 3DPW | PA-MPJPE: 58.0 -> 57.2 | Improved |
| use probabilistic semantic prior instead of manual vertex painting (MVP) (ablation) | DSR · 3DPW | PA-MPJPE: 58.1 -> 57.2 | Improved |
| add DSR-C alone (ablation) | C-EFT · 3DPW | PA-MPJPE: 58.5 -> 57.6 | Improved |
| add DSR-MC alone (s-IoU) (ablation) | C-EFT · 3DPW | PA-MPJPE: 58.5 -> 58.0 | Improved |
| use soft-IoU instead of soft-DistM for DSR-MC (ablation) | DSR-MC · 3DPW | PA-MPJPE: 58.2 -> 58.0 | Improved |
| add DSR loss to standard loss | SD regressor · Human3.6M subset | PA-MPJPE: 47.5 -> 45.1 | Improved |
| add DSR loss to standard loss | SD regressor · Human3.6M subset | PVE: 99.2 -> 96.6 | Improved |
| add DSR loss to standard loss, per joint | SD regressor · Human3.6M subset ankle error | joint error: 99.3 -> 96.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SPIN — SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition (AAAI 2021) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SCALE: Modeling Clothed Humans with a Surface Codec of Articulated Local Elements (CVPR 2021 · similarity 0.91)
- SCULPT: Shape-Conditioned Unpaired Learning of Pose-dependent Clothed and Textured Human Meshes (CVPR 2024 · similarity 0.91)
- SHARP: Shape-Aware Reconstruction of People in Loose Clothing (IJCV 2022 · similarity 0.91)
- 3DPeople: Modeling the Geometry of Dressed Humans (ICCV 2019 · similarity 0.91)
- Learning to Dress 3D People in Generative Clothing (CVPR 2020 · similarity 0.91)
- FreeCloth: Free-form Generation Enhances Challenging Clothed Human Modeling (CVPR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
