Research map / 3D vision / 3D generation
Ins-HOI: Instance Aware Human-Object Interactions Recovery
Jiajun Zhang, Yuxiang Zhang, Hongwen Zhang, Xiao Zhou, Boyao Zhou, Ruizhi Shao, Zonghai Hu, Yebin Liu
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 intersection constraint L_in (ablation) | Ins-HOI · Ins-Sit cross-subject | IoU (%): 8.1% -> 0.058% | Improved |
| add intersection constraint L_in (ablation) | Ins-HOI · Ins-Sit cross-subject | intersection volume (m^3): 8.59e-3 -> 5.86e-5 | Improved |
| add intersection constraint L_in (ablation) | Ins-HOI · Ins-Sit within-subject | IoU (%): 7.65% -> 0.055% | Improved |
| add intersection constraint L_in (ablation) | Ins-HOI · Ins-Sit within-subject | intersection volume (m^3): 8.16e-3 -> 5.56e-5 | Improved |
| add real-captured data to synthetic-only training (ablation) | Ins-HOI · Ins-Sit | reconstruction quality (qualitative) | Improved |
| add reposed synthetic data Syn_R (ablation) | Ins-HOI · Ins-Sit chair | reconstruction quality (qualitative) | Improved |
| add synthetic data to real-captured training (ablation) | Ins-HOI · Ins-Grasp | trainability | Improved |
| raise gamma from 0.5 to 1 (ablation) | Ins-HOI · contact deformation | object deformation (qualitative): 0.5 -> 0.75 -> 1 | Mixed |
| use Ins-HOI instead of PIFu | 3D reconstruction · Ins-Sit cross-subject | Chamfer (cm): 0.5280 -> 0.4828 | Improved |
| use Ins-HOI instead of NeuS2 | 3D reconstruction · Ins-Grasp | Chamfer (cm): 0.2706 -> 0.1535 | Improved |
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.
- NeuS2 — NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction (ICCV 2023) · better on 6 of 6 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
- Reconstructing In-the-Wild Open-Vocabulary Human-Object Interactions (CVPR 2025 · similarity 0.90)
- InterAct: Advancing Large-Scale Versatile 3D Human-Object Interaction Generation (CVPR 2025 · similarity 0.90)
- ContactField: Implicit Field Representation for Multi-Person Interaction Geometry (NeurIPS 2024 · similarity 0.89)
- End-to-End HOI Reconstruction Transformer with Graph-based Encoding (CVPR 2025 · similarity 0.89)
- PICO: Reconstructing 3D People In Contact with Objects (CVPR 2025 · similarity 0.89)
- EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild (CVPR 2025 · similarity 0.89)
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.
