Research map / 3D vision / Human pose estimation
Geometry-Preserving Image Generation for 6D Object Pose Estimation
Jiafeng Zhang, Rui Song, Zhengtai Zhang, Jiaojiao Li, Kailang Cao, Lizhang Peng, David Ferstl, Yinlin Hu
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 ControlNetGeom geometry conditioning (ablation) | GenerationPose · BOP macro avg | NCD (%): 1.318 -> 0.867 | Improved |
| add ControlNetGeom geometry conditioning (ablation) | GenerationPose · BOP macro avg | BF1@2: 0.881 -> 0.898 | Improved |
| add mask-aware attention (ablation) | GenerationPose · BOP macro avg | BF1@2: 0.892 -> 0.898 | Improved |
| add mask-aware attention (ablation) | GenerationPose · BOP macro avg | NCD (%): 0.873 -> 0.867 | Improved |
| add mask dilation (r>0) at inference (ablation) | GenerationPose · BOP macro avg | NCD (%): 0.886 -> 0.867 | Improved |
| use DDIM strength 0.6 instead of 0.3 (ablation) | GenerationPose · BOP macro avg | NCD (%): 1.085 -> 0.867 | Improved |
| use DDIM strength 0.6 instead of 1.0 (ablation) | GenerationPose · BOP macro avg | NCD (%): 1.045 -> 0.867 | Improved |
| translate renders into pseudo-real images | GenerationPose · BOP macro avg (FID) | FID: 132.84 -> 114.57 | Improved |
| translate renders into pseudo-real images | GenerationPose · BOP macro avg (KID) | KID: 0.0986 -> 0.0794 | Improved |
| train WDR on generated data instead of PBR | WDR · BOP LineMOD | ADI.10d: 40.96 -> 79.87 | Improved |
| add generated data to real training data | WDR · BOP LineMOD | ADI.10d: 83.01 -> 85.55 | Improved |
| add generated data rather than PBR to real data | WDR · BOP LineMOD | ADI.10d: 83.54 -> 85.55 | 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.
- GDR-Net — GDR-Net: Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation (CVPR 2021) · better on 3 of 3 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
- Category-Level Articulated Object Pose Estimation via Pose–Shape Hypothesis Generation and Verification (ECCV 2026 · similarity 0.87)
- Can Generative Video Models Help Pose Estimation? (CVPR 2025 · similarity 0.87)
- Reconstruction by Generation: 3D Multi-Object Scene Reconstruction from Sparse Observations (ECCV 2026 · similarity 0.87)
- Sparse-view Pose Estimation and Reconstruction via Analysis by Generative Synthesis (NeurIPS 2024 · similarity 0.86)
- SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects (ICLR 2025 · similarity 0.86)
- Pos3R: 6D Pose Estimation for Unseen Objects Made Easy (CVPR 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.
