AI Paper Atlas

Research map / 3D vision / 3D generation

ReGenHOI: Unifying Reconstruction and Generation for 3D Human-Object Interaction Understanding

Miao Xu, Xiangyu Zhu, Zidu Wang, Xusheng Liang, Bao Li, Jinlin Wu, Zelin Zang, Zhen Lei

CVPR 2026 · 3D vision › 3D generation · approach: Hand-object interaction generation

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 reasoning trace mechanism (RTM) to contact prediction (ablation)ReGenHOI · DAMON human contactF1: 77.0 -> 78.4Improved
add reasoning trace mechanism (RTM) (ablation)ReGenHOI · PICO reconstructionPA-CD human (cm): 6.05 -> 5.42Improved
add gravity-based diffusion bridge (GBDB) refinement (ablation)ReGenHOI · PICO reconstructionInteraction volume IV (cm3): 1.20 -> 0.87Improved
add GBDB refinement (ablation)ReGenHOI · PICO reconstructionPenetration depth PD (mm): 2.45 -> 2.08Improved
add GBDB refinement (ablation)ReGenHOI · PICO reconstructionPA-CD object (cm): 12.84 -> 12.68Mixed
train reconstruction jointly with generation in shared latent space (ablation)ReGenHOI · PICO reconstructionPA-CD human (cm): 5.76 -> 5.42Improved
train reconstruction jointly with generation (ablation)ReGenHOI · PICO reconstructionPA-CD object (cm): 12.15 -> 12.68Hurt
train generation jointly with reconstruction (ablation)ReGenHOI · FullBodyManipulationFrechet Score FS: 0.49 -> 0.44Improved
train generation jointly with reconstruction (ablation)ReGenHOI · FullBodyManipulationFID: 1.03 -> 1.02No clear effect
train generation jointly with reconstruction (ablation)ReGenHOI · BEHAVEFID: 1.42 -> 1.38Improved
reason directly on 3D point clouds instead of 2D liftingReGenHOI · DAMON human contactF1: 75.6 -> 78.4Improved
ReGenHOI vs InteractVLM per object categoryReGenHOI · DAMON semantic contactF1: Food 66.4 -> 66.0; Sports 77.9 -> 78.4Mixed

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