Research map / 3D vision / Human pose estimation
JOTR: 3D Joint Contrastive Learning with Transformers for Occluded Human Mesh Recovery
Jiahao Li, Zongxin Yang, Xiaohan Wang, Jianxin Ma, Chang Zhou, Yi Yang
This paper tried a common change: Add a contrastive loss
add joint-to-joint contrastive loss · 3DPW-OC · MPJPE (mm): 77.0 -> 76.6 Improved
What 1,317 accepted papers report for this change
- 1,253 improved
- 22 no clear effect
- 16 mixed
- 26 hurt
- Beyond Looking Up, Try Looking Around: Harmonizing Global Structure and Local Consistency in Optimal Transport for Short Text Clustering ICML 2026 · add contrastive learning · SearchSnippets · ACC: 56.65 -> 80.12Improved
- SleepLM: Natural-Language Intelligence for Human Sleep ICML 2026 · add contrastive objective to captioning (CoCa vs Cap) · sleep event IoU · IoU: 29.2 -> 27.4Hurt
- LIMSSR: LLM-Driven Sequence-to-Score Reasoning under Training-Time Incomplete Multimodal Observations ICML 2026 · add contrastive loss L_con · FS1000 average · Spearman correlation: 0.759 -> 0.789Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 3D feature sampling to flattened 2D features (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 77.8 -> 77.0 | Improved |
| use flattened 2D features instead of joint-sampled 2D features (no 3D) (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 78.3 -> 77.8 | Improved |
| use flattened 2D features with 3D sampling instead of sampled 2D with 3D sampling (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 77.4 -> 77.0 | Improved |
| add 3D features on top of flattened 2D features (ablation) | JOTR · 3DPW-OC | PA-MPJPE (mm): 53.2 -> 53.6 | Hurt |
| add successive refining layers (coarse-to-fine) instead of initial 2D regression (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 276.5 -> 77.0 | Improved |
| decouple SMPL query tokens from joint query tokens (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 79.8 -> 77.0 | Improved |
| add joint-to-non-joint contrastive loss (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 77.0 -> 75.9 | Improved |
| add joint-to-joint contrastive loss (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 77.0 -> 76.6 | Improved |
| combine both contrastive losses (ablation) | JOTR · 3DPW-OC | MPJPE (mm): 77.0 -> 75.7 | Improved |
| use JOTR instead of 3DCrowdNet | JOTR · 3DPW test | PA-MPJPE (mm): 51.2 -> 48.7 | Improved |
Later papers that compared with this method
1 later accepted paper reports JOTR as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- ScoreHypo: Probabilistic Human Mesh Estimation with Hypothesis Scoring (CVPR 2024 · better on 3 of 3 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
- Instance-aware Contrastive Learning for Occluded Human Mesh Reconstruction (CVPR 2024 · similarity 0.90)
- DPMesh: Exploiting Diffusion Prior for Occluded Human Mesh Recovery (CVPR 2024 · similarity 0.89)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.86)
- ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization (ICML 2025 · similarity 0.86)
- Joint2Human: High-Quality 3D Human Generation via Compact Spherical Embedding of 3D Joints (CVPR 2024 · similarity 0.86)
- FactorizedHMR: A Hybrid Framework for Video Human Mesh Recovery (NeurIPS 2026 · 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.
