AI Paper Atlas

Research map / 3D vision / Human pose estimation

Diff9D: Diffusion-Based Domain-Generalized Category-Level 9-DoF Object Pose Estimation

Jian Liu, Wei Sun, Hui Yang, Pengchao Deng, Chongpei Liu, Nicu Sebe, Hossein Rahmani, Ajmal Saeed Mian

TPAMI 2025 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 29 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 RGB image condition (ablation)Diff9D · REAL2753D75 mAP: 32.9 -> 44.1Improved
add point cloud condition (ablation)Diff9D · REAL2753D75 mAP: 2.9 -> 44.1Improved
add time step condition (ablation)Diff9D · REAL2753D75 mAP: 0.0 -> 44.1Improved
add shape prior condition on top of shape estimator and encoder (ablation)Diff9D · REAL2753D75 mAP: 44.1 -> 43.8No clear effect
add shape estimator and encoder (ablation)Diff9D · REAL2753D75 mAP: 39.3 -> 44.1Improved
add NOCS shape regression and encoding (ablation)Diff9D · REAL2753D75 mAP: 42.7 -> 44.1Improved
use self-attention instead of cross-attention in denoiser (ablation)Diff9D · REAL2753D75 mAP: 16.7 -> 44.1Improved
add skip connections in denoiser (ablation)Diff9D · REAL2753D75 mAP: 41.2 -> 44.1Improved
use 7 transformer blocks instead of 3 (ablation)Diff9D · REAL2755deg2cm mAP: 32.2 -> 36.5Improved
increase transformer blocks beyond 7 (9, 11) (ablation)Diff9D · REAL2755deg2cm mAP: 36.5 -> 36.5No clear effect
use 3 DDIM sampling steps instead of 1 (ablation)Diff9D · REAL2753D75 mAP: 0.0 -> 44.1Improved
increase sampling steps from 3 to 1000 (ablation)Diff9D · REAL2753D75 mAP: 44.1 -> 44.3No clear effect

Later papers that compared with this method

2 later accepted papers report Diff9D 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 1 of 1 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