Supplementary Materials for SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning

One of our main target issues is the semantic drift between true background and future class, and we alleviate the issue using the unknown class modeling with saliency maps.