LightNet: A Lightweight 3D Convolutional Neural Network for Real-Time 3D Object Recognition

Simple item page
dc.contributor.authorZhi, Shuaifengen_US
dc.contributor.authorLiu, Yongxiangen_US
dc.contributor.authorLi, Xiangen_US
dc.contributor.authorGuo, Yulanen_US
dc.contributor.editorIoannis Pratikakis and Florent Dupont and Maks Ovsjanikoven_US
dc.date.accessioned2017-04-22T17:17:40Z
dc.date.available2017-04-22T17:17:40Z
dc.date.issued2017
dc.description.abstractWith the rapid growth of 3D data, accurate and efficient 3D object recognition becomes a major problem. Machine learning methods have achieved the state-of-the-art performance in the area, especially for deep convolutional neural networks. However, existing network models have high computational cost and are unsuitable for real-time 3D object recognition applications. In this paper, we propose LightNet, a lightweight 3D convolutional neural network for real-time 3D object recognition. It achieves comparable accuracy to the state-of-the-art methods with a single model and extremely low computational cost. Experiments have been conducted on the ModelNet and Sydney Urban Objects datasets. It is shown that our model improves the VoxNet model by relative 17.4% and 23.1% on the ModelNet10 and ModelNet40 benchmarks with less than 67% of training parameters. It is also demonstrated that the model can be applied in real-time scenarios.en_US
dc.description.sectionheadersPaper Session I
dc.description.seriesinformationEurographics Workshop on 3D Object Retrieval
dc.identifier.doi10.2312/3dor.20171046
dc.identifier.isbn978-3-03868-030-7
dc.identifier.issn1997-0471
dc.identifier.pages9-16
dc.identifier.urihttps://doi.org/10.2312/3dor.20171046
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/3dor20171046
dc.publisherThe Eurographics Associationen_US
dc.subjectI.3.5 [Computer Graphics]
dc.subjectComputational Geometry and Object Modeling
dc.subjectCurve
dc.subjectsurface
dc.subjectsolid
dc.subjectand object representations
dc.subjectI.4.8 [IMAGE PROCESSING AND COMPUTER VISION]
dc.subjectScene Analysis
dc.subjectShape
dc.subjectI.4.8 [IMAGE PROCESSING AND COMPUTER VISION]
dc.subjectScene Analysis
dc.subjectObject recognition
dc.subjectI.5.1 [PATTERN RECOGNITION]
dc.subjectModels
dc.subjectNeural nets
dc.titleLightNet: A Lightweight 3D Convolutional Neural Network for Real-Time 3D Object Recognitionen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
009-016.pdf
Size:
771.69 KB
Format:
Adobe Portable Document Format
Download
Collections
3DOR 17