Research map / 3D vision / Human pose estimation
CrossInfoNet: Multi-Task Information Sharing Based Hand Pose Estimation
Kuo Du, Xiangbo Lin, Yi Sun, Xiaohong Ma
Check 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 heat-map constraints to the baseline network (ablation) | CrossInfoNet baseline · ICVL | Average 3D distance error (mm): 9.28 -> 9.08 | Improved |
| Add heat-map constraints to the baseline network (ablation) | CrossInfoNet baseline · NYU | Average 3D distance error (mm): 11.17 -> 10.84 | Improved |
| Add cross-connections to the baseline with heat-map constraints (ablation) | CrossInfoNet · ICVL | Average 3D distance error (mm): 9.08 -> 8.48 | Improved |
| Add cross-connections to the baseline with heat-map constraints (ablation) | CrossInfoNet · NYU | Average 3D distance error (mm): 10.84 -> 10.08 | Improved |
| Add heat-map constraints to the cross-connection network (ablation) | CrossInfoNet · ICVL | Average 3D distance error (mm): 8.79 -> 8.48 | Improved |
| Add heat-map constraints to the cross-connection network (ablation) | CrossInfoNet · NYU | Average 3D distance error (mm): 10.57 -> 10.08 | Improved |
| Add cross-connections without heat-map constraints (ablation) | CrossInfoNet baseline · ICVL | Average 3D distance error (mm): 9.28 -> 8.79 | Improved |
| Add cross-connections without heat-map constraints (ablation) | CrossInfoNet baseline · NYU | Average 3D distance error (mm): 11.17 -> 10.57 | Improved |
| Use CrossInfoNet rather than Point-to-Point | ICVL depth-map hand pose estimation · ICVL | Mean error (mm): 6.3 -> 6.73 | Hurt |
| Use CrossInfoNet rather than V2V | NYU hand pose estimation · NYU | Mean error (mm): 8.42 -> 10.08 | Hurt |
| Use CrossInfoNet rather than DenseReg | MSRA hand pose estimation · MSRA | Mean error (mm): 7.2 -> 7.86 | Hurt |
| Use CrossInfoNet rather than DenseReg | MSRA hand pose estimation · MSRA | fps: 27.8 -> 124.5 | Improved |
Later papers that compared with this method
3 later accepted papers report CrossInfoNet as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- HandFoldingNet: A 3D Hand Pose Estimation Network Using Multiscale-Feature Guided Folding of a 2D Hand Skeleton (ICCV 2021 · better on 3 of 3 shared results)
- Hand-Transformer: Non-Autoregressive Structured Modeling for 3D Hand Pose Estimation (ECCV 2020 · better on 3 of 3 shared results)
- JGR-P2O: Joint Graph Reasoning based Pixel-to-Offset Prediction Network for 3D Hand Pose Estimation from a Single Depth Image (ECCV 2020 · better on 2 of 2 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
- Cross-Modal Deep Variational Hand Pose Estimation (CVPR 2018 · similarity 0.87)
- HaMuCo: Hand Pose Estimation via Multiview Collaborative Self-Supervised Learning (ICCV 2023 · similarity 0.87)
- Hand PointNet: 3D Hand Pose Estimation Using Point Sets (CVPR 2018 · similarity 0.87)
- Cross-Domain 3D Hand Pose Estimation With Dual Modalities (CVPR 2023 · similarity 0.87)
- Mining Multi-View Information: A Strong Self-Supervised Framework for Depth-Based 3D Hand Pose and Mesh Estimation (CVPR 2022 · similarity 0.87)
- Crossing Nets: Combining GANs and VAEs With a Shared Latent Space for Hand Pose Estimation (CVPR 2017 · 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.
