Saved in:
Bibliographic Details
Main Authors: Yi, Huahui, Xu, Wei, Qin, Ziyuan, Chen, Xi, Wu, Xiaohu, Li, Kang, Lao, Qicheng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.00406
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918041589121024
author Yi, Huahui
Xu, Wei
Qin, Ziyuan
Chen, Xi
Wu, Xiaohu
Li, Kang
Lao, Qicheng
author_facet Yi, Huahui
Xu, Wei
Qin, Ziyuan
Chen, Xi
Wu, Xiaohu
Li, Kang
Lao, Qicheng
contents Existing prompt-based approaches have demonstrated impressive performance in continual learning, leveraging pre-trained large-scale models for classification tasks; however, the tight coupling between foreground-background information and the coupled attention between prompts and image-text tokens present significant challenges in incremental medical object detection tasks, due to the conceptual gap between medical and natural domains. To overcome these challenges, we introduce the \method~framework, which comprises two main components: 1) Instance-level Prompt Generation (\ipg), which decouples fine-grained instance-level knowledge from images and generates prompts that focus on dense predictions, and 2) Decoupled Prompt Attention (\dpa), which decouples the original prompt attention, enabling a more direct and efficient transfer of prompt information while reducing memory usage and mitigating catastrophic forgetting. We collect 13 clinical, cross-modal, multi-organ, and multi-category datasets, referred to as \dataset, and experiments demonstrate that \method~outperforms existing SOTA methods, with FAP improvements of 5.44\%, 4.83\%, 12.88\%, and 4.59\% in full data, 1-shot, 10-shot, and 50-shot settings, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00406
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle iDPA: Instance Decoupled Prompt Attention for Incremental Medical Object Detection
Yi, Huahui
Xu, Wei
Qin, Ziyuan
Chen, Xi
Wu, Xiaohu
Li, Kang
Lao, Qicheng
Computer Vision and Pattern Recognition
Existing prompt-based approaches have demonstrated impressive performance in continual learning, leveraging pre-trained large-scale models for classification tasks; however, the tight coupling between foreground-background information and the coupled attention between prompts and image-text tokens present significant challenges in incremental medical object detection tasks, due to the conceptual gap between medical and natural domains. To overcome these challenges, we introduce the \method~framework, which comprises two main components: 1) Instance-level Prompt Generation (\ipg), which decouples fine-grained instance-level knowledge from images and generates prompts that focus on dense predictions, and 2) Decoupled Prompt Attention (\dpa), which decouples the original prompt attention, enabling a more direct and efficient transfer of prompt information while reducing memory usage and mitigating catastrophic forgetting. We collect 13 clinical, cross-modal, multi-organ, and multi-category datasets, referred to as \dataset, and experiments demonstrate that \method~outperforms existing SOTA methods, with FAP improvements of 5.44\%, 4.83\%, 12.88\%, and 4.59\% in full data, 1-shot, 10-shot, and 50-shot settings, respectively.
title iDPA: Instance Decoupled Prompt Attention for Incremental Medical Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.00406