Research map / 3D vision / Human pose estimation
CLEP: Contrastive Language-Pose Pretraining
Sen Jia, Huayu Wang, Hsiang-Wei Huang, Zhaochong An, Jenq-Neng Hwang, Huaping Zhang, Lei Li
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 hierarchical (joint-limb-body) pose representation (ablation) | PoseScript-like baseline · CLEP-2M retrieval | mRecall: 64.98 -> 67.90 | Improved |
| add cross-scale attention fusion (SA and GatedHead) (ablation) | CLEP with Hier · CLEP-2M retrieval | mRecall: 67.90 -> 72.79 | Improved |
| switch InfoNCE to SigLIP loss (ablation) | CLEP · CLEP-2M retrieval | mRecall: 72.79 -> 75.69 | Improved |
| pretrain on CLEP-2M with HierFormer, zero-shot to PoseScript-H | CLEP · PoseScript-H zero-shot | mRecall: 5.9 -> 34.8 | Improved |
| train PoseScript baseline on CLEP-2M too (data-controlled comparison) | PoseScript · CLEP-2M test | mRecall: 64.98 -> 75.69 (CLEP) | Improved |
| finetune CLEP on PoseScript vs PoseScript baseline | CLEP · PoseScript-A finetuned | mRecall: 72.8 -> 83.4 | Improved |
| use CLEP encoder for pose generation | PoseScript generator · PoseScript generation | mRecall: 53.2 -> 56.5 | Improved |
| use CLEP encoder for pose editing | PoseFix · PoseFix editing | MPJE: 201 -> 195 (MPJE); Geodesic 8.70 -> 8.79 | Mixed |
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.
- PoseFix — PoseFix: Model-Agnostic General Human Pose Refinement Network (CVPR 2019) · better on 2 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
- UniPose: A Unified Multimodal Framework for Human Pose Comprehension, Generation and Editing (CVPR 2025 · similarity 0.87)
- ChatPose: Chatting about 3D Human Pose (CVPR 2024 · similarity 0.86)
- PoseLLaVA: Pose Centric Multimodal LLM for Fine-Grained 3D Pose Manipulation (AAAI 2025 · similarity 0.86)
- Stable-Pose: Leveraging Transformers for Pose-Guided Text-to-Image Generation (NeurIPS 2024 · similarity 0.86)
- Text as Partial Constraint: Core–Residual Alignment for Robust Vision–Language Learning (NeurIPS 2026 · similarity 0.85)
- HMVLM:Human Motion-Vision-Language Model via MoE LoRA (NeurIPS 2025 · similarity 0.85)
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.
