Research map / 3D vision / NeRF and radiance fields
Transient Neural Radiance Fields for Lidar View Synthesis and 3D Reconstruction
Anagh Malik, Parsa Mirdehghan, Sotiris Nousias, Kyros Kutulakos, David B. Lindell
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 |
|---|---|---|---|
| Use raw photon count histograms (transient supervision) instead of point-cloud depth supervision | Urban NeRF w/mask · Simulated scenes (avg), 5 views | PSNR (dB): 22.34 -> 28.39 | Improved |
| Use transient supervision instead of point-cloud depth supervision | Urban NeRF w/mask · Simulated scenes (avg), 5 views | LPIPS: 0.339 -> 0.115 | Improved |
| Use transient supervision instead of point-cloud depth supervision | Urban NeRF w/mask · Simulated scenes (avg), 5 views | L1 depth: 0.029 -> 0.013 | Improved |
| Use transient supervision instead of point-cloud depth supervision | Urban NeRF w/mask · Captured scenes, 2 views | PSNR (dB): 15.45 -> 22.11 | Improved |
| Use transient supervision instead of point-cloud depth supervision | Urban NeRF w/mask · Captured scenes, 5 views | PSNR (dB): 19.11 -> 22.72 | Improved |
| Use transient supervision instead of point-cloud depth supervision | Urban NeRF w/mask · Captured scenes, 5 views | L1 depth: 0.004 -> 0.010 | Hurt |
| Increase training views from 2 to 5 | Proposed (Transient NeRF) · Simulated scenes (avg) | PSNR (dB): 21.38 -> 28.39 | Improved |
| Increase training views from 2 to 5 | Proposed (Transient NeRF) · Captured scenes | PSNR (dB): 22.11 -> 22.72 | No clear effect |
| Increase training views from 2 to 5 | Proposed (Transient NeRF) · Captured scenes | L1 depth: 0.005 -> 0.010 | Hurt |
| Add depth supervision to Instant NGP (DS-NeRF) | Instant NGP · Captured scenes, 5 views | PSNR (dB): 16.39 -> 14.86 | Hurt |
| Add depth supervision to Instant NGP (DS-NeRF) | Instant NGP · Simulated scenes (avg), 2 views | PSNR (dB): 16.62 -> 19.28 | Improved |
| Add space carving losses to DS-NeRF (Urban NeRF) | DS-NeRF · Simulated scenes (avg), 5 views | L1 depth: 0.119 -> 0.101 | Improved |
Closest papers by meaning
- STGC-NeRF: Spatial-Temporal Geometric Consistency for LiDAR Neural Radiance Fields in Dynamic Scenes (AAAI 2025 · similarity 0.92)
- LiDAR4D: Dynamic Neural Fields for Novel Space-time View LiDAR Synthesis (CVPR 2024 · similarity 0.91)
- NeRF-LiDAR: Generating Realistic LiDAR Point Clouds with Neural Radiance Fields (AAAI 2024 · similarity 0.91)
- LiDAR-RT: Gaussian-based Ray Tracing for Dynamic LiDAR Re-simulation (CVPR 2025 · similarity 0.89)
- PlatoNeRF: 3D Reconstruction in Plato's Cave via Single-View Two-Bounce Lidar (CVPR 2024 · similarity 0.89)
- LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes (CVPR 2024 · 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.
