Newest papers

JointLight: Human-Centric Video Relighting with Asynchronous RGB-Irradiance Joint Modeling

Henglei Lv, Bailin Deng, Xiaoqiang Liu, Yichen Wang, Haoxian Zhang, Pengfei Wan, Lin Gao · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.29780 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Comprender y modelar el transporte de luz en escenas centradas en datos humanos es un desafío de larga data. Mientras que los recientes avances en los modelos de difusión han impulsado la investigación en el relajamiento, el control de iluminación precisa todavía se demora, principalmente debido a la escasez de diversos conjuntos de datos de relajamiento de vídeo de alta calidad, etiquetados con precisión. Los enfoques de relajamiento existentes de extremo a extremo en el transporte de luz se utilizan de forma implícita a través de redes neuronales, lo que no complica sustancialmente la controlabilidad. En contraste, algunos trabajos recurren a pistas de irradiación trazadas como un azulejillo explícito para el transporte de luz. Sin embargo, tales pistas de irradiación construidas son propensas a artefact

Machine translation (Meta M2M100); the English original follows.

Abstract

Understanding and modeling light transport in human-centric scenes is a longstanding challenge. While recent progress in diffusion models has boosted research in relighting, precise illumination control still lags, primarily due to the scarcity of diverse, high-quality, precisely labeled video relighting datasets. Existing end-to-end relighting approaches model light transport implicitly through neural networks, which substantially complicates controllability. In contrast, some works resort to path-traced irradiance cues as an explicit blueprint for light transport. However, such constructed irradiance cues are prone to artifacts and hinder generative quality when directly used as control signals. With the key observation that noise corruption removes high-frequency artifacts and preserves low-frequency light transport, we introduce an asynchronous joint-modeling paradigm that facilitates fine-grained illumination control. During two-stage training, we force the model to jointly denoise RGB and ground-truth irradiance frames with separate timesteps scheduling. A domain matching strategy is applied to the second stage to bridge the domain gap between irradiance cues and ground-truth irradiance maps. At inference time, we introduce an asynchronous hijacked generation pipeline that facilitates flexible illumination control. For training data, we develop a multi-illumination high-quality synthetic video dataset with paired irradiance maps using a powerful game engine. Trained exclusively on this dataset, our model exhibits robust generalization to real- world scenarios and outperforms prior approaches in both lighting fidelity and generative quality. Project page: http://www.geometrylearning.com/JointLight/

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Artificial intelligence, Computer vision, Upsampling, RGB color model, Pipeline (software)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange