Research map / 3D vision / Human pose estimation
Deep Active Contours for Real-time 6-DoF Object Tracking
Long Wang, Shen Yan, Jianan Zhen, Yu Liu, Maojun Zhang, Guofeng Zhang, Xiaowei Zhou
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 |
|---|---|---|---|
| replace handcrafted region statistics with learned boundary prediction (DeepAC vs best optimization-based) | DeepAC · RBOT noisy scenes | 5cm-5deg success rate: 83.2 -> 88.0 | Improved |
| DeepAC on easy RBOT scenes where baselines saturate | DeepAC · RBOT regular/dynamic light/occlusion | 5cm-5deg success rate: 95.2 -> 95.6 (regular) | No clear effect |
| DeepAC versus optimization baselines under strict ADD on BCOT | DeepAC · BCOT | ADD-0.02d: 15.3 -> 24.4 | Improved |
| add statistical foreground and boundary maps as network inputs (ablation) | DeepAC · RBOT | 5cm-5deg: 90.9 -> 93.3 | Improved |
| add multi-level features (ablation) | DeepAC · RBOT | 5cm-5deg: 83.3 -> 93.3 | Improved |
| add learned correspondence line uncertainty (ablation) | DeepAC · BCOT | 5cm-5deg: 92.4 -> 94.0 | Improved |
| add multi-level features (ablation) | DeepAC · BCOT | 2cm-2deg: 51.7 -> 65.5 | Improved |
| increase correspondence lines from 50 to 200 (ablation) | DeepAC · RBOT | 5cm-5deg: 85.8 -> 93.3 | Improved |
| increase correspondence lines from 200 to 300 (ablation) | DeepAC · RBOT | 5cm-5deg: 93.3 -> 93.3 | No clear effect |
| increase line sample range m from 5 to 9 (ablation) | DeepAC · RBOT | 5cm-5deg: 84.1 -> 93.3 | Improved |
| increase line sample range m from 9 to 11 (ablation) | DeepAC · BCOT | 5cm-5deg: 94.0 -> 94.2 | No clear effect |
| DeepAC trained on YCB-V versus render-and-compare learned methods | DeepAC- · RBOT (YCB-V trained) | ADD-0.02d: 10.9 -> 39.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SRT3D — SRT3D: A Sparse Region-Based 3D Object Tracking Approach for the Real World (IJCV 2022) · 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
- Free-Moving Object Reconstruction and Pose Estimation with Virtual Camera (AAAI 2025 · similarity 0.85)
- 6D Object Pose Tracking in Internet Videos for Robotic Manipulation (ICLR 2025 · similarity 0.85)
- FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects (CVPR 2024 · similarity 0.85)
- ProxyPose: 6-DoF Pose Tracking via Video-to-Video Translation (NeurIPS 2026 · similarity 0.85)
- CTO-SLAM: Contour Tracking for Object-Level Robust 4D SLAM (AAAI 2024 · similarity 0.85)
- SingRef6D: Monocular Novel Object Pose Estimation with a Single RGB Reference (NeurIPS 2025 · 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.
