File Download
Supplementary

Conference Paper: TOM-Net: Learning Transparent Object Matting from a Single Image

TitleTOM-Net: Learning Transparent Object Matting from a Single Image
Authors
Issue Date2018
PublisherIEEE Computer Society.
Citation
The 31st IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018), Salt Lake City, UT, 18-22 June 2018 How to Cite?
AbstractThis paper addresses the problem of transparent object matting. Existing image matting approaches for transparent objects often require tedious capturing procedures and long processing time, which limit their practical use. In this paper, we first formulate transparent object matting as a refractive flow estimation problem. We then propose a deep learning framework, called TOM-Net, for learning the refractive flow. Our framework comprises two parts, namely a multi-scale encoder-decoder network for producing a coarse prediction, and a residual network for refinement. At test time, TOM-Net takes a single image as input, and outputs a matte (consisting of an object mask, an attenuation mask and a refractive flow field) in a fast feed-forward pass. As no off-the-shelf dataset is available for transparent object matting, we create a large-scale synthetic dataset consisting of 158K images of transparent objects rendered in front of images sampled from the Microsoft COCO dataset. We also collect a real dataset consisting of 876 samples using 14 transparent objects and 60 background images. Promising experimental results have been achieved on both synthetic and real data, which clearly demonstrate the effectiveness of our approach.
Persistent Identifierhttp://hdl.handle.net/10722/261168

 

DC FieldValueLanguage
dc.contributor.authorChen, G-
dc.contributor.authorHan, K-
dc.contributor.authorWong, KKY-
dc.date.accessioned2018-09-14T08:53:36Z-
dc.date.available2018-09-14T08:53:36Z-
dc.date.issued2018-
dc.identifier.citationThe 31st IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018), Salt Lake City, UT, 18-22 June 2018-
dc.identifier.urihttp://hdl.handle.net/10722/261168-
dc.description.abstractThis paper addresses the problem of transparent object matting. Existing image matting approaches for transparent objects often require tedious capturing procedures and long processing time, which limit their practical use. In this paper, we first formulate transparent object matting as a refractive flow estimation problem. We then propose a deep learning framework, called TOM-Net, for learning the refractive flow. Our framework comprises two parts, namely a multi-scale encoder-decoder network for producing a coarse prediction, and a residual network for refinement. At test time, TOM-Net takes a single image as input, and outputs a matte (consisting of an object mask, an attenuation mask and a refractive flow field) in a fast feed-forward pass. As no off-the-shelf dataset is available for transparent object matting, we create a large-scale synthetic dataset consisting of 158K images of transparent objects rendered in front of images sampled from the Microsoft COCO dataset. We also collect a real dataset consisting of 876 samples using 14 transparent objects and 60 background images. Promising experimental results have been achieved on both synthetic and real data, which clearly demonstrate the effectiveness of our approach.-
dc.languageeng-
dc.publisherIEEE Computer Society.-
dc.relation.ispartofIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)-
dc.rightsIEEE/CVF Conference on Computer Vision and Pattern Recognition. Copyright © IEEE Computer Society.-
dc.rights©2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.-
dc.titleTOM-Net: Learning Transparent Object Matting from a Single Image-
dc.typeConference_Paper-
dc.identifier.emailWong, KKY: kykwong@cs.hku.hk-
dc.identifier.authorityWong, KKY=rp01393-
dc.description.naturepostprint-
dc.identifier.hkuros290286-
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats