Segment new object classes using predictions from related old classes with RaSP
Segment new object classes using predictions from related old classes with RaSP
RaSP: Relation-aware Semantic Prior for Weakly Supervised Incremental Segmentation
arXiv paper abstract https://arxiv.org/abs/2305.19879
arXiv PDF paper https://arxiv.org/pdf/2305.19879.pdf
Class-incremental semantic image segmentation assumes multiple model updates, each enriching the model to segment new categories.
This is typically carried out by providing expensive pixel-level annotations to the training algorithm for all new objects
… image-level labels offer an attractive alternative, yet, such coarse annotations lack precise information about the location and boundary of the new objects.
… argue that, since classes represent not just indices but semantic entities, the conceptual relationships between them can provide valuable information that should be leveraged.
… propose a weakly supervised approach that exploits such semantic relations to transfer objectness prior from the previously learned classes into the new ones, complementing the supervisory signal from image-level labels.
… show how even a simple pairwise interaction between classes can significantly improve the segmentation mask quality of both old and new classes.
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
