Research map / 3D vision / NeRF and radiance fields
ABLE-NeRF: Attention-Based Rendering With Learnable Embeddings for Neural Radiance Field
Zhe Jun Tang, Tat-Jen Cham, Haiyu Zhao
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 |
|---|---|---|---|
| Replace physics-based volumetric rendering with masked attention-based ray rendering | ABLE-NeRF · Blender | PSNR: 33.99 -> 35.02 | Improved |
| Replace physics-based volumetric rendering with masked attention-based ray rendering | ABLE-NeRF · Blender | SSIM: 0.966 -> 0.975 | Improved |
| Replace physics-based volumetric rendering with masked attention-based ray rendering | ABLE-NeRF · Blender | LPIPS: 0.038 -> 0.035 | Improved |
| Use ABLE-NeRF instead of Ref-NeRF | NeRF rendering on Shiny Blender · Shiny Blender | PSNR: 35.96 -> 33.88 | Hurt |
| Use ABLE-NeRF instead of Ref-NeRF | NeRF rendering on Shiny Blender · Shiny Blender | SSIM: 0.967 -> 0.969 | Improved |
| Use ABLE-NeRF instead of Ref-NeRF | NeRF rendering on Shiny Blender · Shiny Blender | LPIPS: 0.058 -> 0.075 | Hurt |
| Add Learnable Embeddings (LE) (ablation) | ABLE-NeRF · Blender | PSNR: 34.05 -> 35.02 | Improved |
| Add Learnable Embeddings (LE) (ablation) | ABLE-NeRF · Blender | SSIM: 0.963 -> 0.975 | Improved |
| Add Learnable Embeddings (LE) (ablation) | ABLE-NeRF · Blender | LPIPS: 0.041 -> 0.035 | Improved |
| Mask rear volumes so each volume cannot attend to those behind it (ablation) | ABLE-NeRF attention renderer · Novel view synthesis | surface rendering accuracy | Improved |
| Append each volume's position on the ray instead of using the masking strategy (ablation) | ABLE-NeRF attention renderer · Novel view synthesis | rendering accuracy | Hurt |
| Add predicted normals (ablation) | Ref-NeRF · Shiny Blender | PSNR: 30.91 -> 35.96 | 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.
- Ref-NeRF — Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fields (CVPR 2022) · better on 4 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
- A Pre-convolved Representation for Plug-and-Play Neural Illumination Fields (AAAI 2024 · similarity 0.91)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.90)
- Aleth-NeRF: Illumination Adaptive NeRF with Concealing Field Assumption (AAAI 2024 · similarity 0.90)
- A General Implicit Framework for Fast NeRF Composition and Rendering (AAAI 2024 · similarity 0.90)
- Inverse Rendering of Glossy Objects via the Neural Plenoptic Function and Radiance Fields (CVPR 2024 · similarity 0.90)
- Physical-aware Neural Radiance Fields for Efficient Exposure Correction (AAAI 2025 · similarity 0.90)
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.
