Revisiting consistency for semi-supervised semantic segmentation

Abstract

Semi-supervised learning an attractive technique in practical deployments of deep models since it relaxes the dependence on labeled data. It is especially important in the scope of dense prediction because pixel-level annotation requires significant effort. This paper considers semi-supervised algorithms that enforce consistent predictions over perturbed unlabeled inputs. We study the advantages of perturbing only one of the two model instances and preventing the backward pass through the unperturbed instance. We also propose a competitive perturbation model as a composition of geometric warp and photometric jittering. We experiment with efficient models due to their importance for real-time and low-power applications. Our experiments show clear advantages of (1) one-way consistency, (2) perturbing only the student branch, and (3) strong photometric and geometric perturbations. Our perturbation model outperforms recent work and most of the contribution comes from photometric component. Experiments with additional data from the large coarsely annotated subset of Cityscapes suggest that semi-supervised training can outperform supervised training with the coarse labels.

The source code is available at https://github.com/Ivan1248/semisup-seg-efficient

References in corpus (6)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Bootstrap your own latent: A new approach to self-supervised Learning
• FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
• Semi-Supervised Medical Image Segmentation via Learning Consistency under Transformations
• Semi-supervised semantic segmentation needs strong, varied perturbations
• Understanding self-supervised Learning Dynamics without Contrastive Pairs

Cited by in corpus (1)

• Semi-Supervised Semantic Segmentation Based on Pseudo-Labels: A Survey