Pushing the efficiency of StereoNet: exploiting spatial sparsity
File(s)
Author(s)
Zampokas, Georgios
Bouganis, Christos-Savvas
Tzovaras, Dimitrios
Type
Conference Paper
Abstract
Current CNN-based stereo matching methods have demonstrated superior performance compared to traditional stereo matching methods. However, mapping these algorithms into embedded devices, which exhibit limited compute resources, and achieving high performance is a challenging task due to the high computational complexity of the CNN-based methods. The recently proposed StereoNet network, achieves disparity estimation with reduced complexity, whereas performance does not greatly deteriorate. Towards pushing this performance to complexity trade-off further, we propose an optimization applied to StereoNet that adapts the computations to the input data, steering the computations to the regions of the input that would benefit from the application of the CNN-based stereo matching algorithm, where the rest of the input is processed by a traditional, less computationally demanding method. Key to the proposed methodology is the introduction of a lightweight CNN that predicts the importance of r efining a region of the input to the quality of the final disparity map, allowing the system to trade-off computational complexity for disparity error on-demand, enabling the application of these methods to embedded systems with real-time requirements.
Editor(s)
Farinella, GM
Radeva, P
Bouatouch, K
Date Issued
2022-01-01
Date Acceptance
2021-12-07
Citation
PROCEEDINGS OF THE 17TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (VISAPP), VOL 5, 2022, pp.757-766
ISSN
2184-4321
Publisher
SCITEPRESS
Start Page
757
End Page
766
Journal / Book Title
PROCEEDINGS OF THE 17TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (VISAPP), VOL 5
Copyright Statement
© 2022 The Author(s). This work is published under CC BY-NC-ND 4.0 International licence.
Identifier
http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000777505000080&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=1ba7043ffcc86c417c072aa74d649202
Source
17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) / 17th International Conference on Computer Vision Theory and Applications (VISAPP)
Subjects
Science & Technology
Technology
Computer Science, Artificial Intelligence
Computer Science, Software Engineering
Imaging Science & Photographic Technology
Computer Science
Deep-learning
Stereo-matching
Sparsity
Publication Status
Published
Start Date
2022-02-06
Finish Date
2022-02-08
Coverage Spatial
ELECTR NETWORK