Research map / 3D vision / Face and shape
HumanOLAT: A Large-Scale Dataset for Full-Body Human Relighting and Novel-View Synthesis
Timo Teufel, Pulkit Gera, Xilong Zhou, Umar Iqbal, Pramod Rao, Jan Kautz, Vladislav Golyanik, Christian Theobalt
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 GS3 instead of PRT-Gaussian | OLAT Gaussian relighting · HumanOLAT 6 subjects | PSNR: 24.06 -> 30.04 | Improved |
| use GS3 instead of BiGS | OLAT Gaussian relighting · HumanOLAT 6 subjects | PSNR: 26.72 -> 30.04 | Improved |
| use RNG instead of GS3 | OLAT Gaussian relighting · HumanOLAT 6 subjects | LPIPS: 0.152 -> 0.139 | Improved |
| use BiGS instead of GS3 | OLAT Gaussian relighting · HumanOLAT 6 subjects | SSIM: 0.892 -> 0.936 | Improved |
| use PRT-Gaussian instead of RNG | OLAT Gaussian relighting · HumanOLAT 6 subjects | LPIPS: 0.139 -> 0.212 | Hurt |
| apply motion compensation (Co-Tracker3 flow with white-light frames) to OLAT frames | image-based relighting · HumanOLAT | visual quality | Improved |
| use Sapiens instead of SAM for masks | mask generation · HumanOLAT | mask accuracy | Improved |
| apply IC-Light to full-body OLAT-lit subjects | IC-Light · HumanOLAT | PSNR: 15.28 -> 17.57 | Hurt |
Closest papers by meaning
- Relightable Holoported Characters: Capturing and Relighting Dynamic Human Performance from Sparse Views (CVPR 2026 · similarity 0.90)
- Relightable Neural Human Assets From Multi-View Gradient Illuminations (CVPR 2023 · similarity 0.90)
- Comprehensive Relighting: Generalizable and Consistent Monocular Human Relighting and Harmonization (CVPR 2025 · similarity 0.88)
- Relighting4D: Neural Relightable Human from Videos (ECCV 2022 · similarity 0.88)
- Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset (CVPR 2025 · similarity 0.86)
- MVHumanNet: A Large-scale Dataset of Multi-view Daily Dressing Human Captures (CVPR 2024 · similarity 0.86)
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.
