ZeroHour
Hugging Face daily paperspublished ()ingested Hejun Wang, Jinxi Li, Junwei Jiang

RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting

infoAI researchimportance 38
AI summary · glm-5.3-flash

RelightFormer is a feed-forward generative transformer for photorealistic single- and multi-view object relighting, trained on a 90K-object dataset.

Researchers introduce RelightFormer, a feed-forward generative transformer adapted from a video foundation model that performs direct image relighting without explicit intrinsic property estimation. The architecture injects target environment maps via a latent illumination module with cross-attention and uses permutation-invariant positional encodings for unordered multi-view inputs. Training relies on the newly constructed Laval Objaverse Dataset (LOD) with 90K objects and 39K unique illuminations, and the model shows state-of-the-art quality with strong zero-shot generalization across single-view, multi-view, and novel-view relighting.

  • Bypasses explicit inverse rendering and intrinsic estimation
  • Latent illumination module injects environment maps via cross-attention
  • Trained on the Laval Objaverse Dataset: 90K objects, 39K illuminations
  • State-of-the-art zero-shot relighting across single- and multi-view tasks
Full article136 words · extracted from huggingface.co · click to collapse

Image relighting is traditionally tackled via complex inverse rendering pipelines, which suffer from ill-posed optimization, or single-image generative models that ignore crucial multi-view cues necessary for understanding 3D geometry and material interactions. To address these limitations, we introduce a feed-forward generative Transformer for direct single- and multi-view image relighting that entirely bypasses explicit intrinsic property estimation. Adapted from a video foundation model, our architecture features a latent illumination module that dynamically injects target environment maps into spatial features via cross-attention. Furthermore, we employ permutation-invariant positional encodings to symmetrically process unordered multi-view inputs without sequential bias. To train this robust data-driven model, we construct the massive Laval Objaverse Dataset (LOD), comprising 90K objects and 39K unique illuminations. Extensive experiments demonstrate state-of-the-art visual quality, photorealistic relighting quality, and strong zero-shot generalization across single-view, multi-view, and novel-view relighting tasks.

Text extracted automatically; images, tables and formatting may be missing. Original: https://huggingface.co/papers/2609.07414