Current studentsExisting staffContact us
 

Structured binary neural networks for accurate image classification and semantic segmentation

Simple item page
dc.contributor.authorZhuang, B.
dc.contributor.authorShen, C.
dc.contributor.authorTan, M.
dc.contributor.authorLiu, L.
dc.contributor.authorReid, I.
dc.contributor.conferenceIEEE Conference on Computer Vision and Pattern Recognition (CVPR) (15 Jun 2019 - 20 Jun 2019 : Long Beach, USA)
dc.date.issued2019
dc.description.abstractIn this paper, we propose to train convolutional neural networks (CNNs) with both binarized weights and activations, leading to quantized models specifically for mobile devices with limited power capacity and computation resources. Previous works on quantizing CNNs seek to approximate the floating-point information using a set of discrete values, which we call value approximation, but typically assume the same architecture as the full-precision networks. However, we take a novel “structure approximation” view for quantization-it is very likely that a different architecture may be better for best performance. In particular, we propose a “network decomposition” strategy, termed Group-Net, in which we divide the network into groups. In this way, each full-precision group can be effectively reconstructed by aggregating a set of homogeneous binary branches. In addition, we learn effective connections among groups to improve the representational capability. Moreover, the proposed Group-Net shows strong generalization to other tasks. For instance, we extend Group-Net for highly accurate semantic segmentation by embedding rich context into the binary structure. Experiments on both classification and semantic segmentation tasks demonstrate the superior performance of the proposed methods over various popular architectures. In particular, we outperform the previous best binary neural networks in terms of accuracy and major computation savings.
dc.description.statementofresponsibilityBohan Zhuang, Chunhua Shen, Mingkui Tan, Lingqiao Liu, Ian Reid
dc.identifier.citationProceedings / CVPR, IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2019, vol.2019-June, pp.413-422
dc.identifier.doi10.1109/CVPR.2019.00050
dc.identifier.isbn9781728132938
dc.identifier.issn1063-6919
dc.identifier.orcidReid, I. [0000-0001-7790-6423]
dc.identifier.urihttp://hdl.handle.net/2440/127246
dc.language.isoen
dc.publisherIEEE
dc.publisher.placeonline
dc.relation.granthttp://purl.org/au-research/grants/arc/DE170101259
dc.relation.ispartofseriesIEEE Conference on Computer Vision and Pattern Recognition
dc.rights©2019 IEEE
dc.source.urihttps://doi.org/10.1109/cvpr.2019.00050
dc.titleStructured binary neural networks for accurate image classification and semantic segmentation
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Computer Science publications