Test-Time Modification: Inverse Domain Transformation for Robust Perception
This addresses robust perception for computer vision systems under environmental shifts, offering a practical alternative to synthetic data generation.
The paper tackles domain generalization by using diffusion models at test time to map target images back to the source distribution where the downstream model was trained, achieving substantial relative gains such as 137% on BDD100K-Night, 68% on ImageNet-R, and 62% on DarkZurich.
Generative foundation models contain broad visual knowledge and can produce diverse image variations, making them particularly promising for advancing domain generalization tasks. While they can be used for training data augmentation, synthesizing comprehensive target-domain variations remains slow, expensive, and incomplete. We propose an alternative: using diffusion models at test time to map target images back to the source distribution where the downstream model was trained. This approach requires only a source domain description, preserves the task model, and eliminates large-scale synthetic data generation. We demonstrate consistent improvements across segmentation, detection, and classification tasks under challenging environmental shifts in real-to-real domain generalization scenarios with unknown target distributions. Our analysis spans multiple generative and downstream models, including an ensemble variant for enhanced robustness. The method achieves substantial relative gains: 137% on BDD100K-Night, 68% on ImageNet-R, and 62% on DarkZurich.