We present the first comprehensive video polyp segmentation(VPS)study in the deep learning era.Over the years,devel-opments in VPS are not moving forward with ease due to the lack of a large-scale dataset with fine-grained segmentation annotations.To address this issue,we first introduce a high-quality frame-by-frame annotated VPS dataset,named SUN-SEG,which contains 158 690 colonoscopy video frames from the well-known SUN-database.We provide additional annotation covering diverse types,i.e.,attribute,object mask,boundary,scribble,and polygon.Second,we design a simple but efficient baseline,named PNS+,which consists of a global encoder,a local encoder,and normalized self-attention(NS)blocks.The global and local encoders receive an anchor frame and multiple successive frames to extract long-term and short-term spatial-temporal representations,which are then progressively refined by two NS blocks.Extensive experiments show that PNS+achieves the best performance and real-time inference speed(170fps),making it a prom-ising solution for the VPS task.Third,we extensively evaluate 13 representative polyp/object segmentation models on our SUN-SEG dataset and provide attribute-based comparisons.Finally,we discuss several open issues and suggest possible research directions for the VPS community.Our project and dataset are publicly available at .

Ge-Peng Ji;Guobao Xiao;Yu-Cheng Chou;Deng-Ping Fan;Kai Zhao;Geng Chen;Luc Van Gool

Research School of Engineering,Australian National University,Canberra 2601,Australia;College of Computer and Control Engineering,Minjiang University,Fuzhou 350108,China;Department of Computer Science,Johns Hopkins University,Baltimore 21218,USA;Computer Vision Laboratory,ETH Zürich,Zürich 8092,Switzerland;Department of Radiological Sciences,University of California,Los Angeles 90095,USA;School of Computer Science and Engineering,Northwestern Polytechnical University,Xi'an 710072,China

机器智能研究（英文）

[1]Ge-Peng Ji;Guobao Xiao;Yu-Cheng Chou;Deng-Ping Fan;Kai Zhao;Geng Chen;Luc Van Gool-.Video Polyp Segmentation:A Deep Learning Perspective)[J].机器智能研究（英文）,2022(06):531-549

scribble,PNS+,PNS+achieves,170fps

Video,Polyp,Segmentation,Deep,Learning,Perspective,We,first,comprehensive,video,polyp,segmentation,VPS,study,deep,learning,Over,years,devel,opments,are,moving,forward,ease,due,lack,large,scale,dataset,grained,annotations,To,address,this,introduce,high,quality,by,annotated,named,SUN,SEG,which,contains,colonoscopy,frames,from,well,known,database,provide,additional,covering,diverse,types,attribute,object,mask,boundary,polygon,Second,design,simple,efficient,baseline,consists,global,local,normalized,self,attention,blocks,encoders,receive,anchor,multiple,successive,extract,long,term,short,spatial,temporal,representations,then,progressively,refined,two,Extensive,experiments,show,that,best,performance,real,inference,speed,making,prom,ising,solution,task,Third,extensively,evaluate,representative,models,our,comparisons,Finally,discuss,several,open,issues,suggest,possible,research,directions,community,Our,project,publicly,available

0.626706