AI Paper Atlas

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

ICCV 2025 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 10 citations (OpenAlex)

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 changedWhereMeasuredEffect
add causal learning (front-door adjustment) to AG-Pose baseline (ablation)AG-Pose · REAL2755°2cm: 57.0 -> 59.7Improved
add knowledge distillation from 3D foundation model (ablation)AG-Pose · REAL2755°2cm: 57.0 -> 57.9Improved
combine causal learning and distillation (ablation)AG-Pose · REAL2755°2cm: 57.0 -> 61.7Improved
add FIFO update to dynamic queue (ablation)CleanPose · REAL2755°2cm: 57.1 -> 61.7Improved
use dynamic queue with FIFO instead of memory bank with similarity update (ablation)CleanPose · REAL2755°2cm: 59.1 -> 61.7Improved
add adaptive fusion in front-door adjustment (ablation)CleanPose · REAL2755°2cm / 10°2cm: 60.5 -> 61.7 (5°2cm), 78.8 -> 78.3 (10°2cm)Mixed
use residual distillation instead of concatenation (ablation)CleanPose · REAL2755°2cm: 60.0 -> 61.7Improved
use PointBert instead of PointNet++ as distillation teacher encoder (ablation)CleanPose · REAL2755°2cm: 59.6 -> 61.7Improved
extract queue initialization from 3D encoder instead of random (ablation)CleanPose · REAL2755°2cm: 58.0 -> 61.7Improved
random feature selection instead of K-means selection (ablation)CleanPose · REAL2755°2cm: 58.5 -> 61.7Improved
set queue length N_q to 80 instead of 20 (ablation)CleanPose · REAL2755°2cm: 57.0 -> 61.7Improved
set sampled features N_s to 12 instead of 80 (ablation)CleanPose · REAL2755°2cm: 56.9 -> 61.7Improved

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.

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy