Learning Reference-Guided Exposure Correction with Hybrid Illumination Characteristics
This work addresses the problem of automatic exposure correction for photographers and computer vision systems, offering a method that does not require ground truth or intrinsic decomposition.
HICNet introduces a reference-guided exposure correction framework that uses illumination embeddings and a hybrid modulation network to adjust exposure while preserving details, achieving state-of-the-art accuracy on public benchmarks and strong generalization to unseen scenes.
We present HICNet, a reference-guided exposure correction framework. A lightweight, content-agnostic encoder distills each image into a compact illumination embedding capturing regional brightness, edge contrast, and higher-order luminance moments. The embedding difference between a source and its reference drives a multi-scale modulation network that combines FiLM-based global adjustment with Photometric Channel Rebalancing for fine-grained, illumination-aware spectral gating, producing exposure-matched outputs while faithfully preserving scene details. A cross-batch contrastive loss orders the illumination manifold, bolstering robustness to diverse lighting conditions. Trained without ground truth or intrinsic decomposition, HICNet attains better accuracy on public benchmarks and generalizes well to entirely unseen scenes.