Research map / 3D vision / Human pose estimation
A2J-Transformer: Anchor-to-Joint Transformer Network for 3D Interacting Hand Pose Estimation From a Single RGB Image
Read the paperCheck 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 |
|---|---|---|---|
| Add Transformer-based anchor refinement to A2J (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 14.44 -> 9.63 | Helped |
| Use anchor-to-joint estimation instead of direct joint regression (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 15.36 -> 9.63 | Helped |
| Predict adaptive 3D anchor weights instead of using uniform weights (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 14.04 -> 9.63 | Helped |
| Use MSDAM instead of original attention (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 10.69 -> 9.63 | Helped |
| Increase depth anchor values from 3 to 7 (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 9.63 -> 9.50 | Helped |
| Increase depth anchor values from 3 to 5 (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 9.63 -> 9.61 | Helped |
| Reduce depth anchor values from 3 to 1 (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 9.63 -> 9.75 | Hurt |
| Reduce in-plane anchor values from 256 to 64 (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 9.63 -> 12.28 | Hurt |
| Reduce in-plane anchor values from 256 to 16 (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 9.63 -> 14.07 | Hurt |
| Reduce in-plane anchor values from 256 to 4 (ablation) | A2J-Transformer · InterHand2.6M | MPJPE (mm): 9.63 -> 15.48 | Hurt |
Closest papers by meaning
- Keypoint Transformer: Solving Joint Identification in Challenging Hands and Object Interactions for Accurate 3D Pose Estimation (CVPR 2022 · similarity 0.90)
- 3D Hand Shape and Pose Estimation From a Single RGB Image (CVPR 2019 · similarity 0.90)
- Model-Based 3D Hand Reconstruction via Self-Supervised Learning (CVPR 2021 · similarity 0.89)
- Keypoint Fusion for RGB-D Based 3D Hand Pose Estimation (AAAI 2024 · similarity 0.89)
- Neural Voting Field for Camera-Space 3D Hand Pose Estimation (CVPR 2023 · similarity 0.88)
- SiMA-Hand: Boosting 3D Hand-Mesh Reconstruction by Single-to-Multi-View Adaptation (AAAI 2024 · similarity 0.87)
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.