DecomposedKnowledgeDistillationfor Class-IncrementalSemanticSegmentation

We introduce a CISS framework that alleviates the forgetting problem and facilitates learning novel classes effectively. We have found that a logit can be decomposed into two terms. They quantify how likely an input belongs toaparticular class ornot, providing aclue forareasoning process ofa model. The KD technique, in this context, preserves the sum of two terms (i.e., a class logit), suggesting that each could be changed and thus the KD does not imitate thereasoning process.