Research map / 3D vision / 3D generation
OneHOI: Unifying Human-Object Interaction Generation and Editing
Jiun Tian Hoe, Weipeng Hu, Xudong Jiang, Yap-Peng Tan, Chee Seng Chan
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 Action Grounding (ablation) | Eligen baseline · HICO-DET HOI generation | HOI accuracy: 0.3061 -> 0.4138 | Improved |
| add Action Grounding (ablation) | Eligen baseline · HICO-DET HOI generation | ImageReward: 0.3921 -> 0.3156 | Hurt |
| add HOI Encoder (ablation) | OneHOI with AG · HICO-DET HOI generation | ImageReward: 0.3156 -> 0.4602 | Improved |
| add HOI Encoder (ablation) | OneHOI with AG · HICO-DET HOI generation | HOI accuracy: 0.4138 -> 0.4254 | Improved |
| add Structured HOI Attention (ablation) | OneHOI with AG+Enc · HICO-DET HOI generation | HOI accuracy: 0.4254 -> 0.4504 | Improved |
| add Structured HOI Attention (ablation) | OneHOI with AG+Enc · MultiHOIEdit | ImageReward: 0.1306 -> 0.1944 | Improved |
| add HOI RoPE (ablation) | OneHOI with AG+Enc+Attn · HICO-DET HOI generation | ImageReward: 0.4861 -> 0.5224 | Improved |
| add HOI RoPE (ablation) | OneHOI with AG+Enc+Attn · MultiHOIEdit | Editability-Identity: 0.433 -> 0.435 | Improved |
| unified training of generation and editing instead of task-specific (ablation) | OneHOI · HICO-DET HOI generation | HOI accuracy: 0.177 -> 0.224 | Improved |
| unified training of generation and editing instead of task-specific (ablation) | OneHOI · IEBench layout-free HOI editing | HOI Editability: 0.464 -> 0.562 | Improved |
| unified training of generation and editing instead of task-specific (ablation) | OneHOI · MultiHOIEdit | HOI Editability: 0.287 -> 0.329 | Improved |
| define action region as union of subject and object instead of between band (ablation) | OneHOI · attention footprint analysis | attention match | Improved |
Closest papers by meaning
- Learning to Generate Human-Human-Object Interactions from Textual Descriptions (NeurIPS 2025 · similarity 0.90)
- CG-HOI: Contact-Guided 3D Human-Object Interaction Generation (CVPR 2024 · similarity 0.90)
- Semantic-Aware Human Object Interaction Image Generation (ICML 2024 · similarity 0.89)
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (CVPR 2025 · similarity 0.89)
- HOIAnimator: Generating Text-prompt Human-object Animations using Novel Perceptive Diffusion Models (CVPR 2024 · similarity 0.89)
- InterAct: Advancing Large-Scale Versatile 3D Human-Object Interaction Generation (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.
