Robust Semantic Segmentation with Superpixel-Mix

Gianni Franchi (1) , Nacim Belkhir , Mai Lan Ha , Yufei Hu (1) , Andrei Bursuc , Volker Blanz , Angela Yao

1 U2IS - Unité d'Informatique et d'Ingénierie des Systèmes

Along with predictive performance and runtime speed, robustness is a key requirement for real-world semantic segmentation. Robustness encompasses accuracy, predictive uncertainty, stability under data perturbation and distribution shift, and reduced bias. To improve robustness, 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 robustness 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.

Domaines: Intelligence artificielle [cs.AI]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Statistiques [math.ST]

Gianni Franchi, Nacim Belkhir, Mai Lan Ha, Yufei Hu, Andrei Bursuc, et al.. Robust Semantic Segmentation with Superpixel-Mix. The British Machine Vision Conference (BMVC), Nov 2021, Online, United Kingdom. ⟨hal-03400621⟩