Transformers,the dominant architecture for natural language processing,have also recently attracted much attention from computational visual media researchers due to their capacity for long-range representation and high performance.Transformers are sequence-to-sequence models,which use a self-attention mechanism rather than the RNN sequential structure.Thus,such models can be trained in parallel and can represent global information.This study comprehensively surveys recent visual transformer works.We categorize them according to task scenario:backbone design,high-level vision,low-level vision and generation,and multimodal learning.Their key ideas are also analyzed.Differing from previous surveys,we mainly focus on visual transformer methods in low-level vision and generation.The latest works on backbone design are also reviewed in detail.For ease of understanding,we precisely describe the main contributions of the latest works in the form of tables.As well as giving quantitative comparisons,we also present image results for low-level vision and generation tasks.Computational costs and source code links for various important works are also given in this survey to assist further development.

Yifan Xu;Huapeng Wei;Minxuan Lin;Yingying Deng;Kekai Sheng;Mengdan Zhang;Fan Tang;Weiming Dong;Feiyue Huang;Changsheng Xu

NLPR,Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China;School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing 100040,China;School of Artificial Intelligence,Jilin University,Changchun 130012,China;Youtu Lab,Tencent Inc.,Shanghai 200233,China;CASIA-LLVISION Joint Lab,Beijing 100190,China

计算可视媒体（英文）

[1]Yifan Xu;Huapeng Wei;Minxuan Lin;Yingying Deng;Kekai Sheng;Mengdan Zhang;Fan Tang;Weiming Dong;Feiyue Huang;Changsheng Xu-.Transformers in computational visual media:A survey)[J].计算可视媒体（英文）,2022(01):33-62

Transformers,computational,visual,media,dominant,architecture,natural,language,processing,have,also,recently,attracted,much,attention,from,researchers,due,their,capacity,long,range,representation,high,performance,are,sequence,models,which,use,self,mechanism,rather,than,RNN,sequential,structure,Thus,such,can,trained,parallel,global,information,This,study,comprehensively,surveys,transformer,works,We,categorize,them,according,scenario,backbone,design,level,vision,low,generation,multimodal,learning,Their,key,ideas,analyzed,Differing,previous,mainly,focus,methods,latest,reviewed,detail,For,ease,understanding,precisely,describe,contributions,tables,well,giving,quantitative,comparisons,image,results,tasks,Computational,costs,source,code,links,various,important,given,this,assist,further,development

0.610058