← Back to all news
arXiv Computer Vision By Yuhan Chen, Wenxuan Yu, Guofa Li, Fuchen Li, Kunyang Huang, Yicui Shi, Ying Fang, Wenbo Chu, Keqiang Li

Gaussian Light Field Splatting: A Physical Prior-Driven Vision Transformer for Low-Light Image Enhancement

Read the original on arXiv Computer Vision →

The paper introduces Gaussian Light Field Splatting (GLFS), a Vision Transformer that embeds continuous light‑field priors derived from 2D Gaussian splatting into its self‑attention mechanism. GLFS models scene illumination as a sum of anisotropic Gaussian basis functions, enabling adaptive spatial gain estimation for low‑light images with complex, non‑uniform lighting. To preserve color fidelity and structural detail, the authors add a color vector angular loss and a luminance edge loss, and report that GLFS outperforms existing methods in illumination correction, structure preservation, and perceptual quality.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

Hugging Face Trending Papers
Aug 11

Towards Color-Faithful Low-Light Image Enhancement via Adaptive Color Debiasing and Saturation Rectification

Low-light imaging often introduces color bias caused by the low signal-to-noise ratio and the image formation process. Although recent low-light image enhancement methods have achieved strong brightness recovery, faithful color restoration remains challenging, manifesting as overall color bias together with local under- and over-saturation.

benchmarkssafety
More like this →
arXiv Computer Vision
Oct 16

Naka-GS: Naka-Guided Chroma Correction and Point-Prior Compaction for Low-Light 3D Gaussian Splatting

Naka-GS is a low‑light 3D Gaussian Splatting framework that jointly enhances observations and improves Gaussian initialization. It introduces Naka‑Guided Chroma Correction, which uses a fixed bionic response as a photometric anchor to learn bounded, frequency‑decoupled luminance and chromatic corrections, preserving fine structures while reducing color and boundary artifacts. A lightweight, non‑learned Point‑Prior Compaction Module refines dense scene priors via voxel pooling and distance‑adaptive pruning, cutting the number of points from 1.00 M to 59.52 K and shortening optimization time by 37.7 % while achieving state‑of‑the‑art PSNR, SSIM, and LPIPS on eight RealX3D scenes.

By Runyu Zhu, SiXun Dong, Zhiqiang Zhang, Qingxia Ye, Rongjun Qin, Zhihua Xu
efficiency
More like this →