{"slug": "lumitokens-3d-relighting-via-token-space-lighting-transformation", "title": "LumiTokens: 3D Relighting via Token-Space Lighting Transformation", "summary": "Researchers introduced LumiTokens, a framework that performs 3D relighting by directly transforming latent scene tokens, avoiding explicit 3D representations and physics-based decomposition. The method, detailed in arXiv:2608.18215v1, encodes lighting as Plucker ray tokens and supports progressive, composable edits, achieving comparable or superior relighting quality to existing methods.", "body_md": "arXiv:2608.18215v1 Announce Type: new\nAbstract: Existing 3D relighting methods operate through either explicit material decomposition, diffusion-based view-space generation, or a combination of both, requiring full recomputation for each new lighting condition. We observe that recent latent scene representations, which encode multi-view images into a set of compact tokens with no fixed physical semantics, open up a novel design space for relighting. We present LumiTokens, a framework that formulates 3D relighting as a direct transformation on latent scene tokens, without explicit 3D representations, rendering equations, or physics-based decomposition. Our model introduces a Scene Token Editor that processes scene tokens jointly with light-ray tokens through self-attention, producing updated tokens that can be decoded into multi-view-consistent relit images. To support diverse lighting types through a unified interface, all lighting signals, including environment maps, point lights, and area lights, are parameterized as Plucker ray tokens, enabling native 3D user interaction with a representation that carries no explicit spatial structure. Crucially, this design supports progressive relighting: because the editor's output remains in the same latent space as its input, a user can incrementally build up illumination one light source at a time, with each edit composing in token space. Experiments demonstrate that LumiTokens achieves comparable or superior relighting quality to other methods and supports progressive, composable lighting edits. Project page: https://neu-vi.github.io/LumiTokens/", "url": "https://wpnews.pro/news/lumitokens-3d-relighting-via-token-space-lighting-transformation", "canonical_source": "https://arxiv.org/abs/2608.18215", "published_at": "2026-08-20 04:00:00+00:00", "updated_at": "2026-08-20 04:13:29.080722+00:00", "lang": "en", "topics": ["artificial-intelligence", "computer-vision", "generative-ai"], "entities": ["LumiTokens", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/lumitokens-3d-relighting-via-token-space-lighting-transformation", "markdown": "https://wpnews.pro/news/lumitokens-3d-relighting-via-token-space-lighting-transformation.md", "text": "https://wpnews.pro/news/lumitokens-3d-relighting-via-token-space-lighting-transformation.txt", "jsonld": "https://wpnews.pro/news/lumitokens-3d-relighting-via-token-space-lighting-transformation.jsonld"}}