Research map / Video / Video segmentation and tracking
Two-Stage Co-Segmentation Network Based on Discriminative Representation for Recovering Human Mesh From Videos
Boyang Zhang, Kehua Ma, Suping Wu, Zhixiang Yuan
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 dual excitation mechanism (DEM) (ablation) | video mesh recovery network · MPI-INF-3DHP | PA-MPJPE: 62.2 -> 61.4 | Improved |
| add fragment superposition in stage 2 (ablation) | video mesh recovery network · MPI-INF-3DHP | PA-MPJPE: 62.5 -> 61.4 | Improved |
| add frequency domain enhancement module (ablation) | video mesh recovery network · MPI-INF-3DHP | PA-MPJPE: 62.3 -> 61.4 | Improved |
| add landmark anchor area loss (area + perimeter) (ablation) | video mesh recovery network · MPI-INF-3DHP | PA-MPJPE: 61.9 -> 61.4 | Improved |
| use only anchor area loss without perimeter, or only perimeter without area (ablation) | landmark anchor loss · MPI-INF-3DHP | PA-MPJPE: 61.8 (area only), 61.7 (perimeter only) vs 61.4 (both) | Mixed |
| segment stage 1 features into 4 or 8 patches instead of 2 (ablation) | stage 1 spatial features · MPI-INF-3DHP | PA-MPJPE: 61.8 (4 patches), 62.8 (8 patches) vs 61.4 | Hurt |
| segment stage 2 into 4 or 8 fragments instead of 2 (ablation) | stage 2 temporal features · MPI-INF-3DHP | Accel: 10.5 (4 fragments), 11.6 (8 fragments) vs 8.5 | Hurt |
| proposed two-stage co-segmentation network vs TCMR | video mesh recovery · 3DPW | MPJPE: 86.5 -> 83.4 | Improved |
| proposed method vs TCMR without 3DPW training data | video mesh recovery · MPI-INF-3DHP (no 3DPW training) | PA-MPJPE: 62.8 -> 61.4 | Improved |
Closest papers by meaning
- Human Mesh Recovery From Monocular Images via a Skeleton-Disentangled Representation (ICCV 2019 · similarity 0.89)
- Human Mesh Recovery From Multiple Shots (CVPR 2022 · similarity 0.88)
- Co-Evolution of Pose and Mesh for 3D Human Body Estimation from Video (ICCV 2023 · similarity 0.88)
- FactorizedHMR: A Hybrid Framework for Video Human Mesh Recovery (NeurIPS 2026 · similarity 0.87)
- MultiPly: Reconstruction of Multiple People from Monocular Video in the Wild (CVPR 2024 · similarity 0.86)
- Cross-view and Cross-pose Completion for 3D Human Understanding (CVPR 2024 · similarity 0.86)
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.
