Research map / Continual learning / Class-incremental

ADAPT: Attentive Self-Distillation and Dual-Decoder Prediction Fusion for Continual Panoptic Segmentation

ICLR 2025 · Continual learning › Class-incremental · approach: Distillation-based continual segmentation

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes ADAPT, a continual panoptic segmentation method combining attentive self-distillation and dual-decoder prediction fusion to reduce catastrophic forgetting when learning a stream of new tasks.

Problem

Continual panoptic segmentation suffers from catastrophic forgetting and poor generalization when learning from a continuous stream of new segmentation tasks.

Approach

Combines attentive self-distillation with a dual-decoder prediction fusion scheme to retain knowledge of earlier tasks while learning new ones in continual panoptic segmentation.

Closest papers by meaning

• CoMFormer: Continual Learning in Semantic and Panoptic Segmentation (CVPR 2023 · similarity 0.91)
• Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation (AAAI 2021 · similarity 0.89)
• ECLIPSE: Efficient Continual Learning in Panoptic Segmentation with Visual Prompt Tuning (CVPR 2024 · similarity 0.89)
• Towards Continual Universal Segmentation (CVPR 2025 · similarity 0.87)
• Mutual Learning for SAM Adaptation: A Dual Collaborative Network Framework for Source-Free Domain Transfer (ICML 2025 · similarity 0.86)
• Beyond Background Shift: Rethinking Instance Replay in Continual Semantic Segmentation (CVPR 2025 · similarity 0.86)