Research map / 3D vision / 3D generation
NeuralHOFusion: Neural Volumetric Rendering Under Human-Object Interactions
Check 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 |
|---|---|---|---|
| combine non-rigid key-volume fusion with neural implicit reconstruction and normal refinement (ablation) | NeuralHOFusion human geometry · Human geometry | P2S×10⁻⁴: 7.7407 -> 2.855 | Helped |
| add normal refinement (ablation) | NeuralHOFusion human geometry · Human geometry | P2S×10⁻⁴: 3.1425 -> 2.855 | Helped |
| add the template-aid object tracking pipeline (ablation) | NeuralHOFusion object geometry · Object geometry | P2S×10⁻⁴: 35.138 -> 11.480 | Helped |
| add normal refinement (ablation) | NeuralHOFusion object geometry · Object geometry | P2S×10⁻⁴: 11.529 -> 11.480 | Helped |
| use layer-wise neural human blending instead of albedo-volume texturing (ablation) | NeuralHOFusion human rendering · Human texture | PSNR: 26.758 -> 30.040 | Helped |
| use layer-wise neural human blending instead of naive neural blending (ablation) | NeuralHOFusion human rendering · Human texture | PSNR: 25.983 -> 30.040 | Helped |
| use temporal neural object blending with non-occluded observations (ablation) | NeuralHOFusion object rendering · Object texture | PSNR: 33.455 -> 37.901 | Helped |
| use temporal neural object blending instead of naive neural blending (ablation) | NeuralHOFusion object rendering · Object texture | PSNR: 30.760 -> 37.901 | Helped |
| increase the input camera count from two to six (ablation) | NeuralHOFusion · Human-object interaction scenes | texture quality | Helped |
| combine the full NeuralHOFusion pipeline rather than using comparison methods | Human-object interaction rendering · Human-object interactions | PSNR: 27.526 -> 33.59 | Helped |
| combine the full NeuralHOFusion pipeline rather than using comparison methods | Human-object interaction geometry · Human-object interactions | P2S×10⁻⁴: 5.770 -> 2.692 | Helped |
Closest papers by meaning
- Instant-NVR: Instant Neural Volumetric Rendering for Human-Object Interactions From Monocular RGBD Stream (CVPR 2023 · similarity 0.91)
- NeuralHumanFVV: Real-Time Neural Volumetric Human Performance Rendering Using RGB Cameras (CVPR 2021 · similarity 0.91)
- Function4D: Real-Time Human Volumetric Capture From Very Sparse Consumer RGBD Sensors (CVPR 2021 · similarity 0.89)
- Real-time Free-view Human Rendering from Sparse-view RGB Videos using Double Unprojected Textures (CVPR 2025 · similarity 0.86)
- Template Free Reconstruction of Human-object Interaction with Procedural Interaction Generation (CVPR 2024 · similarity 0.86)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 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.