Show simple item record

dc.contributor.authorZhang, Jian
dc.contributor.authorZhao, Chen
dc.contributor.authorGao, Wen
dc.date.accessioned2020-03-03T05:41:43Z
dc.date.available2020-03-03T05:41:43Z
dc.date.issued2020-03-02
dc.identifier.citationZhang, J., Zhao, C., & Gao, W. (2020). Optimization-Inspired Compact Deep Compressive Sensing. IEEE Journal of Selected Topics in Signal Processing, 14(4), 765–774. doi:10.1109/jstsp.2020.2977507
dc.identifier.doi10.1109/JSTSP.2020.2977507
dc.identifier.urihttp://hdl.handle.net/10754/661850
dc.description.abstractIn order to improve CS performance of natural images, in this paper, we propose a novel framework to design an OPtimization-INspired Explicable deep Network, dubbed OPINE-Net, for adaptive sampling and recovery. Both orthogonal and binary constraints of sampling matrix are incorporated into OPINE-Net simultaneously. In particular, OPINE-Net is composed of three subnets: sampling subnet, initialization subnet and recovery subnet, and all the parameters in OPINE-Net (\eg sampling matrix, nonlinear transforms, shrinkage threshold, step size) are learned end-to-end, rather than hand-crafted. Moreover, considering the relationship among neighboring blocks, an enhanced version OPINE-${\rm{Net^{+}}}$ is developed, which allows image blocks to be sampled independently but reconstructed jointly to further enhance the performance. In addition, some interesting findings of learned sampling matrix are presented. Compared with existing state-of-the-art network-based CS methods, the proposed hardware-friendly OPINE-Nets not only achieve better performance but also require much fewer parameters and much less storage space, while maintaining a real-time computational speed.
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)
dc.relation.urlhttps://ieeexplore.ieee.org/document/9019857/
dc.relation.urlhttps://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9019857
dc.rightsArchived with thanks to IEEE Journal of Selected Topics in Signal Processing
dc.titleOptimization-Inspired Compact Deep Compressive Sensing
dc.typeArticle
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentVisual Computing Center (VCC)
dc.identifier.journalIEEE Journal of Selected Topics in Signal Processing
dc.eprint.versionPre-print
dc.contributor.institutionSchool of Electronic and Computer Engineering, Peking University Shenzhen Graduate School, 429362 Nanshan, Shenzhen China
dc.contributor.institutionPeking University School of Electronics Engineering and Computer Science, 529005 Beijing China
kaust.personZhao, Chen
refterms.dateFOA2020-03-03T05:42:51Z
dc.date.published-online2020-03-02
dc.date.published-print2020-05


Files in this item

Thumbnail
Name:
09019857.pdf
Size:
4.134Mb
Format:
PDF
Description:
Accepted manuscript

This item appears in the following Collection(s)

Show simple item record