Multi-Frame Feature Aggregation for Real-Time Instrument Segmentation in Endoscopic Video

doi:10.1109/LRA.2021.3096156

CORC > 自动化研究所 > 中国科学院自动化研究所 > 精密感知与控制研究中心 > 精密感知与控制

	Multi-Frame Feature Aggregation for Real-Time Instrument Segmentation in Endoscopic Video
	Lin, Shan 1; Qin, Fangbo 2; Peng, Haonan 1; Bly, Randall A.3; Moe, Kris S.3; Hannaford, Blake 1
刊名	IEEE ROBOTICS AND AUTOMATION LETTERS
	2021-10-01
卷号	6 期号:4 页码:6773-6780
关键词	Computer vision for medical robotics deep learning for visual perception object detection segmentation and categorization
ISSN号	2377-3766
DOI	10.1109/LRA.2021.3096156
通讯作者	Lin, Shan(shanlin0331@gmail.com)
英文摘要	Deep learning-based methods have achieved promising results on surgical instrument segmentation. However, the high computation cost may limit the application of deep models to time-sensitive tasks such as online surgical video analysis for robotic-assisted surgery. Moreover, current methods may still suffer from challenging conditions in surgical images such as various lighting conditions and the presence of blood. We propose a novel Multi-frame Feature Aggregation (MFFA) module to aggregate video frame features temporally and spatially in a recurrent mode. By distributing the computation load of deep feature extraction over sequential frames, we can use a lightweight encoder to reduce the computation costs at each time step. Moreover, public surgical videos usually are not labeled frame by frame, so we develop a method that can randomly synthesize a surgical frame sequence from a single labeled frame to assist network training. We demonstrate that our approach achieves superior performance to corresponding deeper segmentation models on two public surgery datasets.
资助项目	National Science Foundation[IIS-2036255]
WOS研究方向	Robotics
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000678343900013
资助机构	National Science Foundation
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/45640]
专题	精密感知与控制研究中心_精密感知与控制
通讯作者	Lin, Shan
作者单位	1.Univ Washington, Dept Elect & Comp Engn, Seattle, WA 98195 USA 2.Chinese Acad Sci, Res Ctr Precis Sensing & Control, Inst Automat, Beijing 100190, Peoples R China 3.UW, Dept Otolaryngol Head & Neck Surg, Seattle, WA 98105 USA
推荐引用方式 GB/T 7714	Lin, Shan,Qin, Fangbo,Peng, Haonan,et al. Multi-Frame Feature Aggregation for Real-Time Instrument Segmentation in Endoscopic Video[J]. IEEE ROBOTICS AND AUTOMATION LETTERS,2021,6(4):6773-6780.
APA	Lin, Shan,Qin, Fangbo,Peng, Haonan,Bly, Randall A.,Moe, Kris S.,&Hannaford, Blake.(2021).Multi-Frame Feature Aggregation for Real-Time Instrument Segmentation in Endoscopic Video.IEEE ROBOTICS AND AUTOMATION LETTERS,6(4),6773-6780.
MLA	Lin, Shan,et al."Multi-Frame Feature Aggregation for Real-Time Instrument Segmentation in Endoscopic Video".IEEE ROBOTICS AND AUTOMATION LETTERS 6.4(2021):6773-6780.