Research map / 3D vision / Human pose estimation
A Visual Navigation Perspective for Category-Level Object Pose Estimation
Jiaxin Guo, Fangxun Zhong, Rong Xiong, Yun-Hui Liu, Yue Wang, Yiyi Liao
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 a few GD steps after IL_DA (ablation) | IL_DA · Mean over REAL275, Cars, Faces | rotation AP: 67.2 -> 70.0 | Improved |
| add a few GD steps after RL (ablation) | RL · Mean over REAL275, Cars, Faces | rotation AP: 64.2 -> 68.0 | Improved |
| add a few GD steps after IL_DA (ablation) | IL_DA · Mean over REAL275, Cars, Faces | translation AP: 53.9 -> 53.6 | No clear effect |
| add a few GD steps after RL (ablation) | RL · Mean over REAL275, Cars, Faces | translation AP: 50.0 -> 51.4 | Improved |
| replace GD with IL_DA policy | GD · Mean over REAL275, Cars, Faces | rotation AP: 39.7 -> 67.2 | Improved |
| replace GD with RL policy | GD · Mean over REAL275, Cars, Faces | rotation AP: 39.7 -> 64.2 | Improved |
| use IL_DA instead of 32-initial-state GD for speed | IL_DA w/ GD · inference time per image | seconds: 4.68 -> 0.033 | Improved |
| use multi-step inference with on-policy data (IL_DA vs IL_BC) (ablation) | imitation learning · laptop (REAL275, synthetic) | rotation AP | Improved |
| add loss on latent code delta z (ablation) | IL_BC single-step · laptop (REAL275, synthetic) | rotation AP | Improved |
| use GD on a good initialization (10-40 deg offset) (ablation) | GD vs RL/IL · Cars | AP 10/30 deg | Mixed |
Closest papers by meaning
- Generative Category-level Object Pose Estimation via Diffusion Models (NeurIPS 2023 · similarity 0.92)
- GenCape: Structure-Inductive Generative Modeling for Category-Agnostic Pose Estimation (ICLR 2026 · similarity 0.89)
- SchemaPose: RGB-Based Category-Level Pose Estimation with Parametric Category Schema (NeurIPS 2026 · similarity 0.88)
- Sparse-view Pose Estimation and Reconstruction via Analysis by Generative Synthesis (NeurIPS 2024 · similarity 0.88)
- Unsupervised Learning of Category-Level 3D Pose from Object-Centric Videos (CVPR 2024 · similarity 0.87)
- 3D-Aware Hypothesis & Verification for Generalizable Relative Object Pose Estimation (ICLR 2024 · 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.
