Robust Semantic Segmentation with Superpixel-Mix

2021/08/02 by Franchi, Gianni, Belkhir, Nacim, Ha, Mai Lan +4 · 1 citation
#Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2108.00968

Abstract

Along with predictive performance and runtime speed, reliability is a key requirement for real-world semantic segmentation. Reliability encompasses robustness, predictive uncertainty and reduced bias. To improve reliability, we introduce Superpixel-mix, a new superpixel-based data augmentation method with teacher-student consistency training. Unlike other mixing-based augmentation techniques, mixing superpixels between images is aware of object boundaries, while yielding consistent gains in segmentation accuracy. Our proposed technique achieves state-of-the-art results in semi-supervised semantic segmentation on the Cityscapes dataset. Moreover, Superpixel-mix improves the reliability of semantic segmentation by reducing network uncertainty and bias, as confirmed by competitive results under strong distributions shift (adverse weather, image corruptions) and when facing out-of-distribution data.

Cited by

• HSMix: Hard and Soft Mixing Data Augmentation for Medical Image Segmentation

Related

• A Simple Baseline for Semi-supervised Semantic Segmentation with Strong Data Augmentation
• Efficient semantic image segmentation with superpixel pooling
• Adaptive Superpixel for Active Learning in Semantic Segmentation
• Superpixel-based Semantic Segmentation Trained by Statistical Process Control
• Semantic Segmentation with Active Semi-Supervised Representation Learning