Incremental Semantic Segmentation Using Multiple Model Consensus

Class-incremental semantic segmentation (CISS) faces significant challenges due to catastrophic forgetting and background shift problems. While existing approaches typically rely on labeled data to progressively finetune models for new tasks, these methods become impractical when only multiple pre-trained models are available or when labeled data cannot be accessed for either old or new tasks. To address this challenge, we propose Multiple Model Consensus (MMC), which leverages the outputs from multiple pre-trained models on unlabeled data. MMC selectively distills trustworthy knowledge by identifying consensus among the models' predictions. We evaluate our method using Pascal VOC2012 as the training dataset and MS COCO2017 as a source of unlabeled auxiliary data. Experimental results demonstrate the effectiveness of our approach, achieving 6.3% and 15.55% improvement in mean Intersection over Union (mIoU) compared to existing methods after incorporating three consecutive tasks.