GC-ART: Global Learnable Second-Order Rational Tone Curves for Illumination Robustness
For image classification, this provides a simple, efficient method to enhance robustness to illumination changes, though improvements are modest and incremental.
GC-ART introduces a lightweight, differentiable pre-processing module that learns global tone curves from histograms to improve image classification robustness. On CIFAR-10 with ResNet-18, it matches clean accuracy, improves on multiplicative darkening, and achieves 48.45% on contrast corruption (vs. 46.27% baseline and 47.13% Zero-DCE++).
We introduce GC-ART (Global Curve Adaptive Rational Tone-mapping), a lightweight differentiable pre-processing module for robust image classification. GC-ART predicts an endpoint-pinned rational tone curve from per-channel soft histograms using a 643-parameter MLP, then applies the curve pointwise before the classifier. The module is trained end-to-end with cross-entropy and a soft monotonicity penalty. On CIFAR-10 with a CIFAR-style ResNet-18, GC-ART matches clean accuracy with the unenhanced baseline and other learned enhancers, improves over the baseline on multiplicative darkening, and achieves the best learned-method result on contrast corruption (48.45% vs. 46.27% for the baseline and 47.13% for Zero-DCE++). These results suggest that histogram-conditioned rational curves can learn useful global tone corrections, including contrast-expanding behavior, while preserving edge locations by construction through pointwise mapping. GC-ART also uses substantially fewer FLOPs than convolutional learned enhancers at 32 x 32. The current hyperparameters are untuned, leaving room for systematic improvement.