VITA: A Multi-Source Vicinal Transfer Augmentation Method for Out-of-Distribution Generalization
This addresses the need for robust computer vision models against common corruptions, representing an incremental improvement over existing augmentation strategies.
The paper tackles the problem of generating on-manifold augmented samples to improve robustness against diverse image corruptions, proposing VITA, which significantly outperforms state-of-the-art augmentation methods in experiments on corruption benchmarks.
Invariance to diverse types of image corruption, such as noise, blurring, or colour shifts, is essential to establish robust models in computer vision. Data augmentation has been the major approach in improving the robustness against common corruptions. However, the samples produced by popular augmentation strategies deviate significantly from the underlying data manifold. As a result, performance is skewed toward certain types of corruption. To address this issue, we propose a multi-source vicinal transfer augmentation (VITA) method for generating diverse on-manifold samples. The proposed VITA consists of two complementary parts: tangent transfer and integration of multi-source vicinal samples. The tangent transfer creates initial augmented samples for improving corruption robustness. The integration employs a generative model to characterize the underlying manifold built by vicinal samples, facilitating the generation of on-manifold samples. Our proposed VITA significantly outperforms the current state-of-the-art augmentation methods, demonstrated in extensive experiments on corruption benchmarks.