中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
Lightweight real-time stereo matching algorithm for AI chips

文献类型：期刊论文

作者Liu, Yi6; Wang, Wenhao1,2; Xu, Xintao6; Guo, Xiaozhou1,6; Gong, Guoliang1,6; Lu, Huaxiang1,4,5,6
刊名COMPUTER COMMUNICATIONS
出版日期2023-02-01
卷号199页码:210-217
ISSN号0140-3664
关键词AI chips Stereo matching Lightweight network Unsupervised learning Multi-stage stereo matching
DOI10.1016/j.comcom.2022.06.018
英文摘要AI chips have developed rapidly and achieved remarkable acceleration effects in the corresponding algorithm field in recent years. However, deep learning algorithms are changing rapidly, including many operators that AI chips and inference frameworks cannot use in the short term. To solve the problem that it is challenging to deploy a stereo matching algorithm based on binocular vision on AI chips, this paper proposes a multi-stage unsupervised lightweight real-time depth estimation algorithm for AI chips called TradNet. TradNet combines the traditional matching algorithm with a convolutional neural network and uses convolution directly supported by AI chips to realize the structure of the traditional matching algorithm. TradNet is composed of operators directly supported by current AI chips, which reduces the computational complexity of the algorithm, and greatly improves the compatibility of the stereo matching algorithm with existing AI chips. Compared with the deep learning-based multi-stage binocular disparity algorithm AnyNet, the accuracy is improved by 5.12%, and the inference speed is only 12.7%. Compared with the matching-based binocular disparity algorithm BM, the accuracy is improved by 25.24%, and the inference speed is only 48.7%. Our final model can process 1280x720 resolution images within a range of 60-80 FPS on an NVIDIA TITAN Xp. It achieves 28FPS on a 1TOPS (Tera Operations Per Second) custom AI chip, and the power consumption is 0.88 W.
资助项目Beijing Academy of Artificial Intelligence (BAAI) ; CAS Strategic Leading Science and Technology Project[XDA27040303] ; CAS Strategic Leading Science and Technology Project[XDA18040400] ; CAS Strategic Leading Science and Technology Project[XDB44000000] ; National Natural Science Foundation of China[U19A 2080] ; National Natural Science Foundation of China[U1936106] ; High Technology Project[31513070501] ; High Technology Project[1916312ZD0090-2201]
WOS研究方向Computer Science ; Engineering ; Telecommunications
语种英语
出版者ELSEVIER
WOS记录号WOS:000916921000001
源URL[http://119.78.100.204/handle/2XEOYT63/20010]  
专题中国科学院计算技术研究所期刊论文
通讯作者Gong, Guoliang
作者单位1.Univ Chinese Acad Sci, Beijing, Peoples R China
2.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
3.Univ Sci & Technol China, Hefei, Peoples R China
4.Beijing Key Lab Semicond Neural Network Intellige, Beijing, Peoples R China
5.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing, Peoples R China
6.Chinese Acad Sci, Inst Semicond, Beijing, Peoples R China
推荐引用方式
GB/T 7714		 Liu, Yi,Wang, Wenhao,Xu, Xintao,et al. Lightweight real-time stereo matching algorithm for AI chips[J]. COMPUTER COMMUNICATIONS,2023,199:210-217.
APA Liu, Yi,Wang, Wenhao,Xu, Xintao,Guo, Xiaozhou,Gong, Guoliang,&Lu, Huaxiang.(2023).Lightweight real-time stereo matching algorithm for AI chips.COMPUTER COMMUNICATIONS,199,210-217.
MLA Liu, Yi,et al."Lightweight real-time stereo matching algorithm for AI chips".COMPUTER COMMUNICATIONS 199(2023):210-217.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

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。