Do We Need Perfect Data? Leveraging Noise for Domain Generalized Segmentation
By: Taeyeong Kim , SeungJoon Lee , Jung Uk Kim and more
Potential Business Impact:
Teaches computers to understand pictures better, even when messy.
Domain generalization in semantic segmentation faces challenges from domain shifts, particularly under adverse conditions. While diffusion-based data generation methods show promise, they introduce inherent misalignment between generated images and semantic masks. This paper presents FLEX-Seg (FLexible Edge eXploitation for Segmentation), a framework that transforms this limitation into an opportunity for robust learning. FLEX-Seg comprises three key components: (1) Granular Adaptive Prototypes that captures boundary characteristics across multiple scales, (2) Uncertainty Boundary Emphasis that dynamically adjusts learning emphasis based on prediction entropy, and (3) Hardness-Aware Sampling that progressively focuses on challenging examples. By leveraging inherent misalignment rather than enforcing strict alignment, FLEX-Seg learns robust representations while capturing rich stylistic variations. Experiments across five real-world datasets demonstrate consistent improvements over state-of-the-art methods, achieving 2.44% and 2.63% mIoU gains on ACDC and Dark Zurich. Our findings validate that adaptive strategies for handling imperfect synthetic data lead to superior domain generalization. Code is available at https://github.com/VisualScienceLab-KHU/FLEX-Seg.
Similar Papers
Exploring Single Domain Generalization of LiDAR-based Semantic Segmentation under Imperfect Labels
CV and Pattern Recognition
Helps self-driving cars see better in bad weather.
A Framework for Low-Effort Training Data Generation for Urban Semantic Segmentation
CV and Pattern Recognition
Makes fake pictures look like real city photos.
Zero Shot Domain Adaptive Semantic Segmentation by Synthetic Data Generation and Progressive Adaptation
CV and Pattern Recognition
Teaches computers to see new things from descriptions.