NAVE	Networked Augmented Virtual Environment (NAVE) Group

Publication:Maoxian Wan, Kaige Li, Qichuan Geng, Weimin Shi, Zhong Zhou. Incremental Few-Shot Semantic Segmentation via Multi-Level Switchable Visual Prompts. International Conference on Computer Vision 2025 (ICCV'25), Honolulu, Hawaii, United States, Oct 19 – 23th, 2025. (CCF rank-A conference) pdf         Existing incremental few-shot semantic segmentation (IFSS) methods often learn novel classes by fine-tuning parameters from previous stages.  This inevitably reduces the distinguishability of old class features, leading to catastrophic forgetting and overfitting to limited new samples. In this paper, we propose a novel prompt-based IFSS method with a visual prompt pool to store and switch multi-granular knowledge across stages, boosting new class learning capability. Specifically, we introduce three levels of prompts: 1) Task-persistent prompts: capturing generalizable knowledge shared across stages, such as foreground-background distributions, to ensure consistent recognition guidance; 2) Stage-specific prompts: adapting to unique requirements of each stage by integrating its discriminative knowledge (e.g., shape difference) with common knowledge from previous stages; and 3) Region-unique prompts: encoding category-specific structures (e.g., edges) to accurately guide the model to retain local details. In particular, we introduce a prompt switching mechanism that adaptively allocates the knowledge required for base and new classes, avoiding interference between prompts and preventing catastrophic forgetting and reducing increasing computation. Our method achieves new state-of-the-art performance, outperforming previous SoTA methods by 30.28\% mIoU-N on VOC and 13.90\% mIoU-N on COCO under 1-shot. create by admin at 2025-06-26 15:01:30