FSDR: Frequency Space Domain Randomization for Domain Generalization

TL;DR

FSDR introduces a frequency space domain randomization technique that enhances domain generalization by selectively randomizing frequency components, preserving semantic structures and improving segmentation performance across unknown domains.

Contribution

It proposes a novel frequency space randomization method that controls domain-variant features while preserving domain-invariant features, improving domain generalization in segmentation tasks.

Findings

FSDR outperforms existing methods in multiple domain generalizable segmentation tasks.

FSDR achieves performance comparable to domain adaptation methods that use target data during training.

The method effectively preserves semantic structures while randomizing domain-specific features.

Abstract

Domain generalization aims to learn a generalizable model from a known source domain for various unknown target domains. It has been studied widely by domain randomization that transfers source images to different styles in spatial space for learning domain-agnostic features. However, most existing randomization uses GANs that often lack of controls and even alter semantic structures of images undesirably. Inspired by the idea of JPEG that converts spatial images into multiple frequency components (FCs), we propose Frequency Space Domain Randomization (FSDR) that randomizes images in frequency space by keeping domain-invariant FCs (DIFs) and randomizing domain-variant FCs (DVFs) only. FSDR has two unique features: 1) it decomposes images into DIFs and DVFs which allows explicit access and manipulation of them and more controllable randomization; 2) it has minimal effects on semantic structures of images and domain-invariant features. We examined domain variance and invariance property of FCs statistically and designed a network that can identify and fuse DIFs and DVFs dynamically through iterative learning. Extensive experiments over multiple domain generalizable segmentation tasks show that FSDR achieves superior segmentation and its performance is even on par with domain adaptation methods that access target data in training.