Research map / 3D vision / Human pose estimation
CleanPose: Category-Level Object Pose Estimation via Causal Learning and Knowledge Distillation
Xiao Lin, Yun Peng, Liuyi Wang, Xianyou Zhong, Minghao Zhu, Yi Feng, Jingwei Yang, Chengju Liu, Qijun Chen
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 causal learning (front-door adjustment) to AG-Pose baseline (ablation) | AG-Pose · REAL275 | 5°2cm: 57.0 -> 59.7 | Improved |
| add knowledge distillation from 3D foundation model (ablation) | AG-Pose · REAL275 | 5°2cm: 57.0 -> 57.9 | Improved |
| combine causal learning and distillation (ablation) | AG-Pose · REAL275 | 5°2cm: 57.0 -> 61.7 | Improved |
| add FIFO update to dynamic queue (ablation) | CleanPose · REAL275 | 5°2cm: 57.1 -> 61.7 | Improved |
| use dynamic queue with FIFO instead of memory bank with similarity update (ablation) | CleanPose · REAL275 | 5°2cm: 59.1 -> 61.7 | Improved |
| add adaptive fusion in front-door adjustment (ablation) | CleanPose · REAL275 | 5°2cm / 10°2cm: 60.5 -> 61.7 (5°2cm), 78.8 -> 78.3 (10°2cm) | Mixed |
| use residual distillation instead of concatenation (ablation) | CleanPose · REAL275 | 5°2cm: 60.0 -> 61.7 | Improved |
| use PointBert instead of PointNet++ as distillation teacher encoder (ablation) | CleanPose · REAL275 | 5°2cm: 59.6 -> 61.7 | Improved |
| extract queue initialization from 3D encoder instead of random (ablation) | CleanPose · REAL275 | 5°2cm: 58.0 -> 61.7 | Improved |
| random feature selection instead of K-means selection (ablation) | CleanPose · REAL275 | 5°2cm: 58.5 -> 61.7 | Improved |
| set queue length N_q to 80 instead of 20 (ablation) | CleanPose · REAL275 | 5°2cm: 57.0 -> 61.7 | Improved |
| set sampled features N_s to 12 instead of 80 (ablation) | CleanPose · REAL275 | 5°2cm: 56.9 -> 61.7 | Improved |
Later papers that compared with this method
2 later accepted papers report CleanPose as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- GenCOPE: Syn2Real Generalized Category-Level Object Pose Estimation for Robotic Picking (NeurIPS 2026 · better on 5 of 5 shared results)
- ComPose: A Unified Completion-Pose Framework for Robust Category-Level Object Pose Estimation (CVPR 2026 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SecondPose — SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation (CVPR 2024) · better on 3 of 3 shared results
- GCE-Pose — GCE-Pose: Global Context Enhancement for Category-level Object Pose Estimation (CVPR 2025) · 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
- PriorPose: Reference-Guided Joint Deformation and Alignment for Category-Level Object Pose Estimation (ECCV 2026 · similarity 0.89)
- Causal-Inspired Multitask Learning for Video-Based Human Pose Estimation (AAAI 2025 · similarity 0.88)
- SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation (CVPR 2024 · similarity 0.87)
- SchemaPose: RGB-Based Category-Level Pose Estimation with Parametric Category Schema (NeurIPS 2026 · similarity 0.86)
- GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.86)
- GeoReF: Geometric Alignment Across Shape Variation for Category-level Object Pose Refinement (CVPR 2024 · 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.
