Research map / 3D vision / Human pose estimation
Deep Dual Consecutive Network for Human Pose Estimation
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 PTM instead of summing previous, current, and next frame heatmaps (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.0 -> 82.8 | Helped |
| Add PTM instead of using only the current-frame heatmap (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.2 -> 82.8 | Helped |
| Add Pose Residual Fusion (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.1 -> 82.8 | Helped |
| Use five dilation rates rather than one (d=3) (ablation) | DCPose Pose Correction Network · PoseTrack2017 validation | mean AP: 81.7 -> 82.8 | Helped |
| Use five dilation rates rather than two (d∈{3,6}) (ablation) | DCPose Pose Correction Network · PoseTrack2017 validation | mean AP: 81.9 -> 82.8 | Helped |
| Use five dilation rates rather than three (d∈{3,6,9}) (ablation) | DCPose Pose Correction Network · PoseTrack2017 validation | mean AP: 82.7 -> 82.8 | Helped |
| Use five dilation rates rather than four (d∈{3,6,9,12}) (ablation) | DCPose Pose Correction Network · PoseTrack2017 validation | mean AP: 82.6 -> 82.8 | Helped |
| Add previous-frame input to PTM and PRF (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.2 -> 82.8 | Helped |
| Add next-frame input to PTM and PRF (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.3 -> 82.8 | Helped |
| Increase temporal interval from T=1 to T=2 (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.8 -> 82.6 | Hurt |
| Increase temporal interval from T=1 to T=3 (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.8 -> 82.1 | Hurt |
| Increase temporal interval from T=1 to T=4 (ablation) | DCPose · PoseTrack2017 validation | mean AP: 82.8 -> 81.7 | Hurt |
Closest papers by meaning
- Temporal Feature Alignment and Mutual Information Maximization for Video-Based Human Pose Estimation (CVPR 2022 · similarity 0.92)
- 3D Human Pose Estimation Using Spatio-Temporal Networks with Explicit Occlusion Training (AAAI 2020 · similarity 0.91)
- Learning Temporal Pose Estimation from Sparsely-Labeled Videos (NeurIPS 2019 · similarity 0.91)
- Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videos (AAAI 2021 · similarity 0.90)
- Video-Based Human Pose Regression via Decoupled Space-Time Aggregation (CVPR 2024 · similarity 0.89)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (AAAI 2025 · similarity 0.89)
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.