One Weird Trick to Improve Your Semi-Weakly Supervised Semantic Segmentation Model
One Weird Trick to Improve Your Semi-Weakly Supervised Semantic Segmentation Model
Wonho Bae, Junhyug Noh, Milad Jalali Asadabadi, Danica J. Sutherland
Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence
Main Track. Pages 2805-2811.
https://doi.org/10.24963/ijcai.2022/389
Semi-weakly supervised semantic segmentation (SWSSS) aims to train a model to identify objects in images based on a small number of images with pixel-level labels, and many more images with only image-level labels. Most existing SWSSS algorithms extract pixel-level pseudo-labels from an image classifier - a very difficult task to do well, hence requiring complicated architectures and extensive hyperparameter tuning on fully-supervised validation sets. We propose a method called prediction filtering, which instead of extracting pseudo-labels, just uses the classifier as a classifier: it ignores any segmentation predictions from classes which the classifier is confident are not present. Adding this simple post-processing method to baselines gives results competitive with or better than prior SWSSS algorithms. Moreover, it is compatible with pseudo-label methods: adding prediction filtering to existing SWSSS algorithms further improves segmentation performance.
Keywords:
Machine Learning: Weakly Supervised Learning
Computer Vision: Segmentation
Machine Learning: Semi-Supervised Learning