Research map / Video / Video generation
Light-X: Generative 4D Video Rendering with Camera and Illumination Control
Tianqi Liu, Zhaoxi Chen, Zihao Huang, Shaocong Xu, Saining Zhang, Chongjie Ye, Bohan Li, Zhiguo Cao, Wei Li, Hao Zhao, Ziwei Liu
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 static-scene training data (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 123.35 -> 101.06 | Improved |
| add dynamic-scene training data (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 108.70 -> 101.06 | Improved |
| add AI-generated training data (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 102.09 -> 101.06 | Improved |
| add fine-grained lighting cues (projected relit views) (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 143.02 -> 101.06 | Improved |
| add global lighting control (Light-DiT) (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 103.13 -> 101.06 | Improved |
| use projected relit cues instead of light-text concatenation (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 137.05 -> 101.06 | Improved |
| use real footage as ground truth instead of algorithm-generated (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 137.83 -> 101.06 | Improved |
| relight a single frame instead of all frames (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: FID 71.10 vs 101.06; Motion Pres. 4.238 vs 2.007 | Mixed |
| add soft illumination masks (ablation) | Light-X · Joint camera-illumination control (ablation) | FID: 148.51 -> 101.06 | Improved |
| Light-X vs TL-Free (joint control) | Light-X · Joint camera-illumination control | FID: 122.73 -> 101.06 | Improved |
| Light-X vs LAV (text relighting) | Light-X · Text-conditioned video relighting | FID: 112.45 -> 83.65 | 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.
- Light-A-Video — Light-A-Video: Training-free Video Relighting via Progressive Light Fusion (ICCV 2025) · 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
- UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting (NeurIPS 2025 · similarity 0.91)
- IllumiCraft: Unified Geometry and Illumination Diffusion for Controllable Video Generation (NeurIPS 2025 · similarity 0.90)
- SyncLight: Single-Edit Multi-View Relighting (NeurIPS 2026 · similarity 0.90)
- TC-Light: Temporally Coherent Generative Rendering for Realistic World Transfer (NeurIPS 2025 · similarity 0.90)
- UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback (NeurIPS 2025 · similarity 0.90)
- LightIt: Illumination Modeling and Control for Diffusion Models (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.
