Revisiting consistency for semi-supervised semantic segmentation
This work addresses the challenge of reducing reliance on labeled data for dense prediction tasks like semantic segmentation, which is important for real-time and low-power applications, but it is incremental as it builds on existing consistency-based methods.
The paper tackled the problem of semi-supervised semantic segmentation by proposing a method that enforces consistency over perturbed unlabeled inputs, showing advantages such as one-way consistency and strong perturbations, with experiments indicating that semi-supervised training can outperform supervised training using coarse labels.
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.