RaSP: Relation-aware Semantic Prior for Weakly Supervised Incremental Segmentation
This work addresses the practical limitation of requiring costly pixel-level annotations in incremental segmentation, offering a weakly supervised alternative that leverages semantic relationships, though it is incremental as it builds on existing weakly supervised methods.
The paper tackles the problem of class-incremental semantic image segmentation by proposing a weakly supervised method that uses semantic relations between classes to transfer objectness prior, reducing the need for expensive pixel-level annotations. The result shows significant improvement in segmentation mask quality for both old and new classes, validated across continual learning tasks, including longer sequences and few-shot scenarios.
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, limiting the adoption of such methods in practical applications. Approaches that solely require image-level labels offer an attractive alternative, yet, such coarse annotations lack precise information about the location and boundary of the new objects. In this paper we argue that, since classes represent not just indices but semantic entities, the conceptual relationships between them can provide valuable information that should be leveraged. We 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. We validate our approach on a number of continual learning tasks, and show how even a simple pairwise interaction between classes can significantly improve the segmentation mask quality of both old and new classes. We show these conclusions still hold for longer and, hence, more realistic sequences of tasks and for a challenging few-shot scenario.