Research map / 3D vision / 3D generation
AV-NeRF: Learning Neural Fields for Real-World Audio-Visual Scene Synthesis
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 |
|---|---|---|---|
| remove AV-Mapper (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.791 | Hurt |
| remove AV-Mapper (ablation) | AV-NeRF · RWAVS overall | ENV: 0.145 -> 0.150 | Hurt |
| remove coordinate transformation (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.701 | Hurt |
| remove coordinate transformation (ablation) | AV-NeRF · RWAVS overall | ENV: 0.145 -> 0.149 | Hurt |
| concatenate visual features to input instead of adding (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.507 | No clear effect |
| add visual information to all A-NeRF layers instead of input only (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.505 | No clear effect |
| absolute direction with positional encoding instead of relative embedding (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.701 | Hurt |
| relative direction with positional encoding instead of relative embedding (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.508 | No clear effect |
| remove both position and direction inputs (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.822 | Hurt |
| remove direction input only (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.688 | Hurt |
| remove position input only (ablation) | AV-NeRF · RWAVS overall | MAG: 1.504 -> 1.817 | Hurt |
| AV-NeRF vs strongest baseline ViGAS | AV-NeRF · RWAVS overall | MAG: 1.830 -> 1.504 | Helped |
Closest papers by meaning
- AV-GS: Learning Material and Geometry Aware Priors for Novel View Acoustic Synthesis (NeurIPS 2024 · similarity 0.91)
- NeRAF: 3D Scene Infused Neural Radiance and Acoustic Fields (ICLR 2025 · similarity 0.90)
- AE-NeRF: Audio Enhanced Neural Radiance Field for Few Shot Talking Head Synthesis (AAAI 2024 · similarity 0.88)
- Real Acoustic Fields: An Audio-Visual Room Acoustics Dataset and Benchmark (CVPR 2024 · similarity 0.88)
- Read, Watch and Scream! Sound Generation from Text and Video (AAAI 2025 · similarity 0.88)
- SoundBrush: Sound as a Brush for Visual Scene Editing (AAAI 2025 · similarity 0.88)
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.