Research map / 3D vision / Depth and stereo
Rendering-Aware HDR Environment Map Prediction from a Single Image
Jun-Peng Xu, Chenyu Zuo, Fang-Lue Zhang, Miao 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 |
|---|---|---|---|
| use the proposed two-stage method instead of the strongest relighting baseline | HDR environment map prediction · Laval Indoor HDR test set, relighting | RMSE: 0.0583 -> 0.0494 | Improved |
| use the proposed two-stage method instead of the strongest relighting baseline | HDR environment map prediction · Laval Indoor HDR test set, relighting | si-RMSE: 0.0583 -> 0.0420 | Improved |
| use the proposed two-stage method instead of the strongest relighting baseline | HDR environment map prediction · Laval Indoor HDR test set, relighting | Angular Error: 10.31 -> 7.35 | Improved |
| use the proposed two-stage method instead of the strongest environment-map baseline | HDR environment map prediction · Laval Indoor HDR test set, environment map | RMSE: 0.2117 -> 0.1877 | Improved |
| use the proposed two-stage method instead of the strongest environment-map baseline | HDR environment map prediction · Laval Indoor HDR test set, environment map | si-RMSE: 0.1539 -> 0.1411 | Improved |
| use the proposed two-stage method instead of the strongest environment-map baseline | HDR environment map prediction · Laval Indoor HDR test set, environment map | Angular Error: 30.23 -> 27.20 | Improved |
| add logarithmic intensity loss to GMLight (ablation) | DLA-SK SG regression · Laval Indoor HDR | RME: 1.385 -> 1.120 | Improved |
| add sum loss alongside logarithmic intensity loss (ablation) | DLA-SK SG regression · Laval Indoor HDR | RME: 1.120 -> 1.233 | Hurt |
| add Transformer to DLA-SK (ablation) | SG regression with GMLight loss · Laval Indoor HDR | RME: 1.385 -> 1.041 | Improved |
| add Transformer and logarithmic intensity loss to DLA-SK (ablation) | SG regression · Laval Indoor HDR | RME: 1.120 -> 0.639 | Improved |
| add Transformer and sum loss to DLA-SK with logarithmic loss (ablation) | SG regression · Laval Indoor HDR | RME: 1.233 -> 0.363 | Improved |
| use coarse-to-fine SH coefficient weighting instead of standard coefficient MSE (ablation) | SH regression · Laval Indoor HDR, rendered objects | RMSE: 0.0452 -> 0.0437 | Improved |
Closest papers by meaning
- HDR Environment Map Estimation for Real-Time Augmented Reality (CVPR 2021 · similarity 0.92)
- High Dynamic Range Novel View Synthesis with Single Exposure (ICML 2025 · similarity 0.89)
- LuxDiT: Lighting Estimation with Video Diffusion Transformer (NeurIPS 2025 · similarity 0.88)
- Perceptual Assessment and Optimization of HDR Image Rendering (CVPR 2024 · similarity 0.87)
- HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splatting (NeurIPS 2024 · similarity 0.87)
- GaussHDR: High Dynamic Range Gaussian Splatting via Learning Unified 3D and 2D Local Tone Mapping (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.
