Knowledge Commons of Institute of Automation,CAS
Enhancing Class-incremental Object Detection in Remote Sensing Through Instance-aware Distillation | |
Feng HT(冯航涛)1,2; Zhang L(张璐)1,2; Yang X(杨旭)1,2; Liu ZY(刘智勇)1,2,3 | |
发表期刊 | Neurocomputing |
ISSN | 0925-2312 |
2024 | |
卷号 | 583页码:127552 |
通讯作者 | Liu, Zhiyong(zhiyong.liu@ia.ac.cn) |
文章类型 | 期刊论文 |
摘要 | Object detection plays a important role within the field of remote sensing, boasting significant applications including intelligent monitoring and urban planning. However, traditional models are constrained by predefined classes and encounter a challenge known as catastrophic forgetting when attempting to learn new classes post-deployment. To address this problem, we propose a novel Instance-aware Distillation approach for Class-incremental Object Detection (IDCOD). Our approach capitalizes on the teacher model, a model from a previous stage, to serve as a guide during the training of the new model on novel data. This methodology facilitates the gradual acquisition of knowledge about new classes while simultaneously preserving the performance achieved on previously learned classes. Instance-aware distillation with masks of old and new classes aims to reduce forgetting and impact on new classes. Furthermore, we design a pseudo-label module to expand old class training data. Experiments on the challenging DOTA dataset, DIOR dataset, RTDOD dataset and PASCAL VOC dataset show that our method effectively detects old classes, incrementally detects new classes, and mitigates catastrophic forgetting. |
关键词 | Class-incremental object detection Remote sensing Object detection |
DOI | 10.1016/j.neucom.2024.127552 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | National Key Research and Development Plan of China[2020AAA0108902] ; National Natural Science Foundation (NSFC) of China[61973301] ; National Natural Science Foundation (NSFC) of China[61972020] ; National Natural Science Foundation (NSFC) of China[62206288] ; Youth Innovation Promotion Association CAS |
项目资助者 | National Key Research and Development Plan of China ; National Natural Science Foundation (NSFC) of China ; Youth Innovation Promotion Association CAS |
WOS研究方向 | Computer Science |
WOS类目 | Computer Science, Artificial Intelligence |
WOS记录号 | WOS:001226169700001 |
出版者 | ELSEVIER |
七大方向——子方向分类 | 目标检测、跟踪与识别 |
国重实验室规划方向分类 | 视觉信息处理 |
是否有论文关联数据集需要存交 | 否 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/56532 |
专题 | 多模态人工智能系统全国重点实验室 |
通讯作者 | Liu ZY(刘智勇) |
作者单位 | 1.中国科学院大学人工智能学院 2.中国科学院自动化研究所 3.Nanjing Artificial Intelligence Research of IA |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Feng HT,Zhang L,Yang X,et al. Enhancing Class-incremental Object Detection in Remote Sensing Through Instance-aware Distillation[J]. Neurocomputing,2024,583:127552. |
APA | Feng HT,Zhang L,Yang X,&Liu ZY.(2024).Enhancing Class-incremental Object Detection in Remote Sensing Through Instance-aware Distillation.Neurocomputing,583,127552. |
MLA | Feng HT,et al."Enhancing Class-incremental Object Detection in Remote Sensing Through Instance-aware Distillation".Neurocomputing 583(2024):127552. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Class-incremental Ob(2152KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论