Research map / Architectures and efficiency / Attention and transformers
Pose Recognition With Cascade Transformers
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| increase input size from 384 × 288 to 512 × 384 | PRTR with ResNet-50 on COCO val2017 · COCO val2017 | AP: 68.2 -> 71.0 | Helped |
| increase input size from 384 × 288 to 512 × 384 | PRTR with ResNet-101 on COCO val2017 · COCO val2017 | AP: 70.1 -> 72.0 | Helped |
| increase query count from 17 to 50 or 100 (ablation) | PRTR on COCO val2017 · COCO val2017 | AP: 67.3 -> 67.6; 67.3 -> 67.7 | No clear effect |
| use non-class-specific, dynamically assigned queries instead of fixed class-specific queries (ablation) | PRTR on COCO val2017 · COCO val2017 | AP: 56.3 -> 67.3 | Helped |
| use ground-truth cropping boxes instead of predicted boxes (ablation) | PRTR on COCO val2017 · COCO val2017 | AP: 67.1 -> 69.1 | Helped |
| include background logits during inference rather than excluding them (ablation) | PRTR on COCO val2017 · COCO val2017 | AP: 67.1 -> 66.2 | Hurt |
| apply flip test (ablation) | PRTR on COCO val2017 · COCO val2017 | AP: 67.1 -> 67.7 | Helped |
Closest papers by meaning
- Towards Accurate Facial Landmark Detection via Cascaded Transformers (CVPR 2022 · similarity 0.87)
- Cascade Transformers for End-to-End Person Search (CVPR 2022 · similarity 0.86)
- Bottom-Up Human Pose Estimation via Disentangled Keypoint Regression (CVPR 2021 · similarity 0.85)
- SDPose: Tokenized Pose Estimation via Circulation-Guide Self-Distillation (CVPR 2024 · similarity 0.84)
- Meta-Point Learning and Refining for Category-Agnostic Pose Estimation (CVPR 2024 · similarity 0.83)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (AAAI 2025 · similarity 0.83)
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.