Research map / 3D vision / Depth and stereo
InvRGB+L: Inverse Rendering of Complex Scenes with Unified Color and LiDAR Reflectance Modeling
Xiaoxue Chen, Bhargav Chandaka, Chih-Hao Lin, Ya-Qin Zhang, David Forsyth, Hao Zhao, Shenlong Wang
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 |
|---|---|---|---|
| add LiDAR-to-RGB albedo consistency loss (ablation) | InvRGB+L relighting · Relighting | PSNR: 29.97 -> 30.42 | Improved |
| add LiDAR-to-RGB albedo consistency loss (ablation) | InvRGB+L relighting · Relighting | LPIPS: 0.34 -> 0.30 | Improved |
| add LiDAR-to-RGB albedo consistency loss (ablation) | InvRGB+L relighting · Relighting | SSIM: 0.73 -> 0.72 | No clear effect |
| add LiDAR input (initialization and albedo consistency) to RGB-only inverse rendering (ablation) | InvRGB+L · Waymo dynamic scene PBR reconstruction | PSNR: 33.35 -> 34.76 | Improved |
| add dynamic scene graph nodes (ablation) | InvRGB+L · Waymo dynamic scene PBR reconstruction | PSNR: 29.13 -> 34.76 | Improved |
| add dynamic scene graph nodes (ablation) | InvRGB+L · Waymo dynamic scene PBR reconstruction | LPIPS: 0.21 -> 0.11 | Improved |
| add specular term to LiDAR reflectance (PBR instead of Lambertian) (ablation) | InvRGB+L · LiDAR intensity ablation (one scene) | Intensity RMSE: 0.0493 -> 0.0470 | Improved |
| add RGB-X diffusion albedo prior (ablation) | InvRGB+L · Albedo estimation (qualitative) | albedo smoothness, shadow ambiguity | Improved |
| compare against LiDAR simulation baselines on intensity | InvRGB+L · Waymo LiDAR intensity NVS | Intensity RMSE avg: 0.073 -> 0.063 | Improved |
| fine-tune YOLOv5 with simulated nighttime images from daytime logs | YOLOv5 · Waymo night detection | mAP@50: 0.236 -> 0.321 | 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.
- AlignMiF — AlignMiF: Geometry-Aligned Multimodal Implicit Field for LiDAR-Camera Joint Synthesis (CVPR 2024) · better on 1 of 1 shared result
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
- Neural Inverse Rendering from Propagating Light (CVPR 2025 · similarity 0.88)
- Inverse Rendering using Multi-Bounce Path Tracing and Reservoir Sampling (ICLR 2025 · similarity 0.88)
- Latent Intrinsics Emerge from Training to Relight (NeurIPS 2024 · similarity 0.87)
- GI-GS: Global Illumination Decomposition on Gaussian Splatting for Inverse Rendering (ICLR 2025 · similarity 0.87)
- ReCap: Better Gaussian Relighting with Cross-Environment Captures (CVPR 2025 · similarity 0.87)
- Blurred LiDAR for Sharper 3D: Robust Handheld 3D Scanning with Diffuse LiDAR and RGB (CVPR 2025 · similarity 0.87)
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.
