Research map / Architectures and efficiency / CNN design
CRF-CNN: Modeling Structured Information in Human Pose Estimation
Xiao Chu, Wanli Ouyang, hongsheng Li, Xiaogang Wang
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 |
|---|---|---|---|
| use CRF-CNN instead of Chu et al. | VGG pose estimator · LSP | mean PCP: 81.1 -> 83.1 | Improved |
| use CRF-CNN instead of Yang et al. | VGG pose estimator · LSP | mean PCP: 81.1 -> 83.1 | Improved |
| use CRF-CNN instead of Chen&Yuille | VGG pose estimator · LSP | mean PCP: 75.0 -> 83.1 | Improved |
| use CRF-CNN instead of Chen and Yuille | pose estimator · FLIC | PCK@0.2 elbow: 95.3 -> 98.0 | Improved |
| use CRF-CNN instead of Chen and Yuille | pose estimator · FLIC | PCK@0.2 wrist: 92.4 -> 94.1 | Improved |
| pass messages twice instead of once with flooding (ablation) | CRF-CNN tree · LSP | mean PCP: 76.6 -> 77.1 | Improved |
| add extra connections between joints (loopy graph) to 2-iter flooding (ablation) | CRF-CNN · LSP | mean PCP: 77.1 -> 78.4 | Improved |
| replace flooding with serial message passing (ReLU) (ablation) | CRF-CNN tree · LSP | mean PCP: 77.1 -> 80.1 | Improved |
| replace ReLU with scaled softmax message nonlinearity (ablation) | CRF-CNN serial tree · LSP | mean PCP: 80.1 -> 83.1 | Improved |
| scale softmax with alpha=0.5 and beta=N_c (ablation) | CRF-CNN · training convergence | convergence speed | Improved |
| serial instead of flooding message passing (ablation) | CRF-CNN · LSP | mean PCP: 78.4 -> 83.1 | Improved |
Closest papers by meaning
- End-To-End Learning of Deformable Mixture of Parts and Deep Convolutional Neural Networks for Human Pose Estimation (CVPR 2016 · similarity 0.86)
- Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimation (IJCV 2016 · similarity 0.86)
- Deep Globally Constrained MRFs for Human Pose Estimation (ICCV 2017 · similarity 0.86)
- Adversarial PoseNet: A Structure-Aware Convolutional Network for Human Pose Estimation (ICCV 2017 · similarity 0.85)
- Thin-Slicing Network: A Deep Structured Model for Pose Estimation in Videos (CVPR 2017 · similarity 0.85)
- Structured Feature Learning for Pose Estimation (CVPR 2016 · 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.
