Research map / 3D vision / Human pose estimation
Putting People in Their Place: Affordance-Aware Human Insertion Into Scenes
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use the proposed method instead of Stable Diffusion | Person hallucination · Person hallucination | FID: 19.651 -> 8.390 | Helped |
| use the proposed method instead of Stable Diffusion | Person hallucination · Person hallucination | PCKh: 0.023 -> 23.213 | Helped |
| use the proposed method instead of Stable Diffusion | Scene hallucination · Scene hallucination | FID: 44.687 -> 20.268 | Helped |
| add video training rather than image-only training (ablation) | Conditional generation · Conditional generation | PCKh: 8.321 -> 15.797 | Helped |
| add video training rather than image-only training (ablation) | Conditional generation · Conditional generation | FID: 13.174 -> 12.103 | Helped |
| add data augmentation to image-only training (ablation) | Conditional generation · Conditional generation | PCKh: 8.321 -> 10.660 | Helped |
| add data augmentation to image-only training (ablation) | Conditional generation · Conditional generation | FID: 13.174 -> 13.008 | Helped |
| use CLIP ViT-L/14 embeddings instead of concatenated VAE features (ablation) | Conditional generation · Conditional generation | FID: 14.956 -> 10.078 | Helped |
| use CLIP ViT-L/14 embeddings instead of concatenated VAE features (ablation) | Conditional generation · Conditional generation | PCKh: 13.020 -> 17.602 | Helped |
| use a larger 860M UNet instead of a smaller UNet (ablation) | Conditional generation · Conditional generation | FID: 12.366 -> 11.232 | Helped |
| use a larger 860M UNet instead of a smaller UNet (ablation) | Conditional generation · Conditional generation | PCKh: 15.095 -> 15.873 | Helped |
| initialize the large UNet with Stable Diffusion weights (ablation) | Conditional generation · Conditional generation | FID: 11.232 -> 10.078 | Helped |
Closest papers by meaning
- Move as You Say Interact as You Can: Language-guided Human Motion Generation with Scene Affordance (CVPR 2024 · similarity 0.87)
- IAAO: Interactive Affordance Learning for Articulated Objects in 3D Environments (CVPR 2025 · similarity 0.87)
- HOComp: Interaction-Aware Human-Object Composition (NeurIPS 2025 · similarity 0.87)
- Vision-Guided Action: Enhancing 3D Human Motion Prediction with Gaze-informed Affordance in 3D Scenes (CVPR 2025 · similarity 0.86)
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (CVPR 2025 · similarity 0.86)
- Weakly-Supervised Affordance Grounding Guided by Part-Level Semantic Priors (ICLR 2025 · similarity 0.86)
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.