Segment Anything is A Good Pseudo-label Generator for Weakly Supervised Semantic Segmentation
This work addresses the long-standing challenge of semantic segmentation with weak labels, but it is incremental as it applies an existing model to a new task.
The authors tackled the problem of weakly supervised semantic segmentation by using the Segment Anything model as a pseudo-label generator, achieving improved results on the PASCAL VOC 2012 dataset.
Weakly supervised semantic segmentation with weak labels is a long-lived ill-posed problem. Mainstream methods mainly focus on improving the quality of pseudo labels. In this report, we attempt to explore the potential of 'prompt to masks' from the powerful class-agnostic large segmentation model, segment-anything. Specifically, different weak labels are used as prompts to the segment-anything model, generating precise class masks. The class masks are utilized to generate pseudo labels to train the segmentation networks. We have conducted extensive experiments on PASCAL VOC 2012 dataset. Experiments demonstrate that segment-anything can serve as a good pseudo-label generator. The code will be made publicly available.