CVJun 22

C^2GR: Coupled Comprehensive Generative Replay for a Continually Learnable Universal Segmentation Model

arXiv:2606.234738.9Has Code
Predicted impact top 57% in CV · last 90 daysOriginality Incremental advance
AI Analysis

For continual learning in medical image segmentation, this work provides a privacy-preserving method to mitigate forgetting across tasks with simultaneous shifts in modality and segmentation objective.

The paper tackles catastrophic forgetting in universal segmentation models under concurrent appearance and objective shifts across sequentially arriving medical tasks. The proposed C^2GR framework achieves only a 2.44% drop in overall performance compared to joint training on 20 diverse tasks.

Universal segmentation models exhibit significant potential for diverse tasks involving different imaging modalities and segmentation objectives. Task-Incremental Learning provides a privacy-preserving approach to continually evolve a universal model on tasks from sequentially-arriving medical departments. However, training the model solely on the incoming task induces forgetting on past tasks, since consecutive tasks exhibit concurrent shifts in image appearance and segmentation objective. To address this problem, we propose a novel Coupled Comprehensive Generative Replay (C^2GR) framework that simultaneously synthesizes image-mask pairs of previous tasks to mitigate forgetting under concurrent appearance and objective shifts. This requires preserving image-mask correspondence for structure-realistic generation and bridging asynchronous optimization of the generator and segmentor for segmentation-oriented generation. Specifically, we propose a Bayesian Joint Diffusion (BJD) method that formulates the correspondence as conditional distributions optimized via conditional denoising. Furthermore, we develop a Relation-aware Unified Prompt Synchronization (RUPS) scheme to simultaneously modulate the generator and segmentor via a shared task-relation-aware prompt for synchronizing their optimization. Experiments on 20 tasks spanning diverse modalities and objectives demonstrate that C^2GR exhibits only a 2.44% drop in overall performance compared to joint training with all task data, effectively alleviating forgetting from the concurrent shifts. Our code will be made publicly available at https://github.com/mar-cry/C2GR.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes