Preventing Catastrophic Forgetting through Memory Networks in Continuous Detection

arXiv:2403.14797v1 Announce Type: cross
Abstract: Modern pre-trained architectures struggle to retain previous information while undergoing continuous fine-tuning on new tasks. Despite notable progress in continual classification, systems designed for complex vision tasks such as detection or segmentation still struggle to attain satisfactory performance. In this work, we introduce a memory-based detection transformer architecture to adapt a pre-trained DETR-style detector to new tasks while preserving knowledge from previous tasks. We propose a novel localized query function for efficient information retrieval from …

abstract architectures arxiv catastrophic forgetting classification continual continuous cs.cv cs.lg detection fine-tuning information memory modern networks performance progress segmentation struggle systems tasks through type vision work