n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation
This work improves segmentation accuracy for computer vision applications, but it is incremental as it builds on an existing method.
The paper tackles semi-supervised semantic segmentation by generalizing cross pseudo supervision to n networks, achieving new state-of-the-art results on Pascal VOC 2012 and Cityscapes datasets across multiple supervised regimes.
We present n-CPS - a generalisation of the recent state-of-the-art cross pseudo supervision (CPS) approach for the task of semi-supervised semantic segmentation. In n-CPS, there are n simultaneously trained subnetworks that learn from each other through one-hot encoding perturbation and consistency regularisation. We also show that ensembling techniques applied to subnetworks outputs can significantly improve the performance. To the best of our knowledge, n-CPS paired with CutMix outperforms CPS and sets the new state-of-the-art for Pascal VOC 2012 with (1/16, 1/8, 1/4, and 1/2 supervised regimes) and Cityscapes (1/16 supervised).