Research map / 3D vision / Human pose estimation
End-to-End Detection and Pose Estimation of Two Interacting Hands
Dong Uk Kim, Kwang In Kim, Seungryul Baek
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 GAN-type interacting hand pose discriminator loss (ablation) | Ours · InterHand2.6M (interacting hands only) | MPJPE (mm): 12.39 -> 11.52 | Improved |
| add GAN-type interacting hand pose discriminator loss (ablation) | Ours · InterHand2.6M (entire test set) | MPJPE (mm): 12.17 -> 12.08 | Improved |
| add joint visibility estimation and visibility-guided heatmap enhancement (ablation) | Ours (-LD) · InterHand2.6M (interacting hands only) | MPJPE (mm): 12.95 -> 12.39 | Improved |
| add joint visibility estimation and visibility-guided heatmap enhancement (ablation) | Ours (-LD) · InterHand2.6M (entire test set) | MPJPE (mm): 12.23 -> 12.17 | Improved |
| add joint training of the two interacting hands (interacting class) (ablation) | Ours (-LV,LD) · InterHand2.6M (interacting hands only) | MPJPE (mm): 14.36 -> 12.95 | Improved |
| add joint training of the two interacting hands (interacting class) (ablation) | Ours (-LV,LD) · InterHand2.6M (entire test set) | MPJPE (mm): 12.39 -> 12.23 | Improved |
| train hand detection and pose estimation end-to-end instead of a separately frozen detector (ablation) | Ours · InterHand2.6M (entire test set) | MPJPE (mm): 13.69 -> 12.08 | Improved |
| train hand detection and pose estimation end-to-end instead of a separately frozen detector (ablation) | Ours · InterHand2.6M (interacting hands only) | MPJPE (mm): 12.48 -> 11.52 | Improved |
| use full end-to-end interacting-hand model versus InterNet baseline | Ours vs Moon et al. · InterHand2.6M | MPJPE (mm): 12.58 -> 12.08 | Improved |
| use full end-to-end interacting-hand model versus best prior method | Ours vs Wei et al. · Ego3D | 2D EPE (px): 7.11 -> 4.53 | Improved |
| use full end-to-end interacting-hand model versus best prior method | Ours vs Moon et al. · Ego3D | 3D EPE (mm): 12.20 -> 11.63 | Improved |
| use full end-to-end interacting-hand model versus best prior method | Ours vs Boukhayma et al. · Tzionas | 2D EPE (px): 12.91 -> 12.42 | Improved |
Closest papers by meaning
- InterHand2.6M: A Dataset and Baseline for 3D Interacting Hand Pose Estimation from a Single RGB Image (ECCV 2020 · similarity 0.92)
- Robust Hand Pose Estimation During the Interaction With an Unknown Object (ICCV 2017 · similarity 0.91)
- Crossing Nets: Combining GANs and VAEs With a Shared Latent Space for Hand Pose Estimation (CVPR 2017 · similarity 0.91)
- Keypoint Transformer: Solving Joint Identification in Challenging Hands and Object Interactions for Accurate 3D Pose Estimation (CVPR 2022 · similarity 0.91)
- Depth-Based 3D Hand Pose Estimation: From Current Achievements to Future Goals (CVPR 2018 · similarity 0.91)
- Weakly-Supervised Domain Adaptation via GAN and Mesh Model for Estimating 3D Hand Poses Interacting Objects (CVPR 2020 · similarity 0.90)
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.
