AIGS-Net: Compact Illumination Field Modeling via 2D Gaussian Splatting for Fast Low-Light Image Enhancement

2026-06-16

AI Summary

This work addresses the trade-off between representational capacity and computational complexity in illumination modeling for low-light image enhancement by proposing AIGS-Net, an ultra-lightweight network. The method introduces, for the first time, input-adaptive 2D Gaussian splatting to construct the illumination field, combined with ordered alpha compositing to achieve spatially varying illumination compensation. It further incorporates a zero-parameter multi-scale contextual encoding module to guide the enhancement process. With only approximately 40 learnable parameters, AIGS-Net integrates noise mask estimation, single-channel gamma mapping, and constraints enforcing cross-channel consistency and target color alignment. Evaluated on the LOL and LSRW benchmarks, the model achieves significant improvements in detail recovery and color fidelity while maintaining extremely low parameter count and high inference efficiency.

Abstract

Existing low-light image enhancement methods often face a bottleneck between the representation capacity of illumination-field modeling and computational complexity. To address this issue, this paper proposes an Adaptive Illumination Gaussian Splatting Network (AIGS-Net), an ultra-lightweight architecture for fast low-light enhancement. Unlike conventional static priors, AIGS-Net constructs an input-adaptive 2D Gaussian Splatting illumination field. The opacity of Gaussian basis functions is dynamically modulated by relative luminance statistics of the input image, and spatially varying illumination compensation is rendered through ordered alpha compositing. To guide adaptive illumination compensation efficiently, a zero-parameter nonlinear multiscale contextual encoding module is introduced to extract low-frequency structures and local contrast cues without additional convolutional weights. To suppress noise amplification and sensor-induced color bias, AIGS-Net integrates noise-mask estimation, locked single-channel Gamma mapping, cross-channel consistency regularization, and target color-alignment constraints. Experiments on LOL and LSRW benchmarks show that AIGS-Net improves detail recovery and color fidelity while requiring only approximately 40 learnable parameters, achieving an effective trade-off between enhancement quality and extreme inference efficiency.

Problem

• low-light image enhancement
• illumination field modeling
• computational complexity
• detail recovery
• color fidelity

Innovation

• 2D Gaussian Splatting
• adaptive illumination modeling
• ultra-lightweight network
• zero-parameter encoding
• low-light enhancement