Research map / 3D vision / Human pose estimation
IST-Net: Prior-Free Category-Level Pose Estimation with Implicit Space Transformation
Jianhui Liu, Yukang Chen, Xiaoqing Ye, Xiaojuan Qi
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 |
|---|---|---|---|
| add implicit space transformation supervision (IST) (ablation) | IST-Net baseline · REAL275 | 5°5cm: 42.9 -> 48.5 | Improved |
| add camera-space enhancer (ablation) | IST-Net with IST · REAL275 | 5°5cm: 48.5 -> 52.5 | Improved |
| add world-space enhancer (ablation) | IST-Net with IST · REAL275 | 3D75: 72.2 -> 75.9 | Improved |
| add position encoding term (ablation) | IST-Net · REAL275 | 5°5cm: 47.4 -> 53.4 | Improved |
| replace category-specific shape prior with random noise (ablation) | DPDN · REAL275 | 3D75: 76.0 -> 76.1 | No clear effect |
| replace category-specific shape prior with random noise (ablation) | SGPA · REAL275 | 3D75: 61.9 -> 60.3 | No clear effect |
| remove prior deformation module | DPDN · REAL275 | accuracy | Improved |
| use implicit instead of explicit space transformation (ablation) | IST-Net · REAL275 | FPS: 22 -> 34 | Improved |
| use implicit instead of explicit space transformation (ablation) | IST-Net · REAL275 | 3D75: 75.1 -> 73.9 | Mixed |
| use MSE instead of L1 for the feature loss (ablation) | IST-Net · REAL275 | 5°5cm: 50.4 -> 53.4 | Improved |
| set feature loss weight lambda_f to 10 instead of 1 or 100 (ablation) | IST-Net · REAL275 | 5°5cm: 47.0 -> 53.4 | Improved |
| train IST-Net without fine-tuning on Wild6D | IST-Net · Wild6D | 3D50: 68.2 -> 79.6 | Improved |
Later papers that compared with this method
3 later accepted papers report IST-Net as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Towards Real-World Aerial Vision Guidance With Categorical 6D Pose Tracker (TPAMI 2025 · better on 3 of 3 shared results)
- U-COPE: Taking a Further Step to Universal 9D Category-level Object Pose Estimation (ECCV 2024 · better on 1 of 1 shared result)
- Omni6DPose: A Benchmark and Model for Universal 6D Object Pose Estimation and Tracking (ECCV 2024 · better on 1 of 1 shared result)
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
- SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation (CVPR 2024 · similarity 0.89)
- CatFormer: Category-Level 6D Object Pose Estimation with Transformer (AAAI 2024 · similarity 0.88)
- Unsupervised Learning of Category-Level 3D Pose from Object-Centric Videos (CVPR 2024 · similarity 0.88)
- 3D-Aware Hypothesis & Verification for Generalizable Relative Object Pose Estimation (ICLR 2024 · similarity 0.88)
- Universal Features Guided Zero-Shot Category-Level Object Pose Estimation (AAAI 2025 · similarity 0.87)
- SchemaPose: RGB-Based Category-Level Pose Estimation with Parametric Category Schema (NeurIPS 2026 · 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.
