Research map / 3D vision / Human pose estimation
Thin-Slicing Network: A Deep Structured Model for Pose Estimation in Videos
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 |
|---|---|---|---|
| add spatial inference to the fully convolutional network (ablation) | Thin-Slicing Network · Penn Action test | PCK@0.2 mean: 87.0 -> 91.4 | Helped |
| add spatio-temporal inference to the spatial-inference model (ablation) | Thin-Slicing Network · Penn Action test | PCK@0.2 mean: 91.4 -> 96.5 | Helped |
| jointly train the ConvNet and inference layers end-to-end rather than keeping ConvNet weights fixed (ablation) | Thin-Slicing Network · Penn Action test | PCK@0.2 mean: 92.8 -> 96.5 | Helped |
| add end-to-end training of the ConvNet with inference layers, then use the ConvNet without inference (ablation) | Thin-Slicing Network · Penn Action test | PCK@0.2 mean: 87.0 -> 91.0 | Helped |
| extend temporal connections across 2 frames instead of 1 (ablation) | Thin-Slicing Network · Penn Action test | PCK@0.2 mean: 96.5 -> 96.4 | No clear effect |
| add spatial inference to the fully convolutional network (ablation) | Thin-Slicing Network · JHMDB | PCK@0.2 mean: 70.9 -> 76.5 | Helped |
| add spatio-temporal inference to the spatial-inference model (ablation) | Thin-Slicing Network · JHMDB | PCK@0.2 mean: 76.5 -> 81.6 | Helped |
| add spatial inference to the fully convolutional network (ablation) | Thin-Slicing Network · sub-JHMDB | PCK@0.2 mean: 82.3 -> 86.9 | Helped |
| add spatio-temporal inference to the spatial-inference model (ablation) | Thin-Slicing Network · sub-JHMDB | PCK@0.2 mean: 86.9 -> 92.1 | Helped |
| use spatio-temporal inference rather than spatial inference only (ablation) | Thin-Slicing Network · JHMDB | PCK@0.1 mean: 60.2 -> 68.7 | Helped |
Closest papers by meaning
- Thin-Slicing for Pose: Learning to Understand Pose Without Explicit Pose Estimation (CVPR 2016 · similarity 0.90)
- Deep Dual Consecutive Network for Human Pose Estimation (CVPR 2021 · similarity 0.90)
- Sparseness Meets Deepness: 3D Human Pose Estimation From Monocular Video (CVPR 2016 · similarity 0.89)
- Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videos (AAAI 2021 · similarity 0.88)
- 3D Human Pose Estimation Using Spatio-Temporal Networks with Explicit Occlusion Training (AAAI 2020 · similarity 0.88)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (AAAI 2025 · similarity 0.86)
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.