Robustness via Uncertainty-aware Cycle Consistency
This work improves robustness in unpaired image translation for applications like autonomous driving and medical imaging, but it is incremental as it builds on existing cycle consistency methods.
The paper tackles the problem of unpaired image-to-image translation by addressing robustness to outliers and predictive uncertainty, proposing a probabilistic method that models per-pixel residuals with a generalized Gaussian distribution. Experimental results show the method exhibits stronger robustness to unseen perturbations in test data across various datasets, including natural images and medical imaging.
Unpaired image-to-image translation refers to learning inter-image-domain mapping without corresponding image pairs. Existing methods learn deterministic mappings without explicitly modelling the robustness to outliers or predictive uncertainty, leading to performance degradation when encountering unseen perturbations at test time. To address this, we propose a novel probabilistic method based on Uncertainty-aware Generalized Adaptive Cycle Consistency (UGAC), which models the per-pixel residual by generalized Gaussian distribution, capable of modelling heavy-tailed distributions. We compare our model with a wide variety of state-of-the-art methods on various challenging tasks including unpaired image translation of natural images, using standard datasets, spanning autonomous driving, maps, facades, and also in medical imaging domain consisting of MRI. Experimental results demonstrate that our method exhibits stronger robustness towards unseen perturbations in test data. Code is released here: https://github.com/ExplainableML/UncertaintyAwareCycleConsistency.