Show simple item record

dc.contributor.authorZhao, Hang
dc.contributor.authorGan, Chuang
dc.contributor.authorRouditchenko, Andrew
dc.contributor.authorVondrick, Carl Martin
dc.contributor.authorMcDermott, Joshua Hartman
dc.contributor.authorTorralba, Antonio
dc.date.accessioned2020-01-20T20:55:35Z
dc.date.available2020-01-20T20:55:35Z
dc.date.issued2018-10-06
dc.identifier.isbn9783030012458
dc.identifier.isbn9783030012465
dc.identifier.issn0302-9743
dc.identifier.issn1611-3349
dc.identifier.urihttps://hdl.handle.net/1721.1/123480
dc.description.abstractWe introduce PixelPlayer, a system that, by leveraging large amounts of unlabeled videos, learns to locate image regions which produce sounds and separate the input sounds into a set of components that represents the sound from each pixel. Our approach capitalizes on the natural synchronization of the visual and audio modalities to learn models that jointly parse sounds and images, without requiring additional manual supervision. Experimental results on a newly collected MUSIC dataset show that our proposed Mix-and-Separate framework outperforms several baselines on source separation. Qualitative results suggest our model learns to ground sounds in vision, enabling applications such as independently adjusting the volume of sound sources. Keywords: Cross-modal learning; Sound separation and localizationen_US
dc.description.sponsorshipNational Science Foundation (U.S.) (Grant IIS-1524817)en_US
dc.language.isoen
dc.publisherSpringer Natureen_US
dc.relation.isversionofhttp://dx.doi.org/10.1007/978-3-030-01246-5_35en_US
dc.rightsCreative Commons Attribution-Noncommercial-Share Alikeen_US
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/4.0/en_US
dc.sourcearXiven_US
dc.titleThe Sound of Pixelsen_US
dc.typeBooken_US
dc.identifier.citationZhao, Hang et al. "The Sound of Pixels." Computer Vision – European Conference on Computer Vision (ECCV 2018), September 4-18, 2018, Munich, Germany, edited by V. Ferrari, M. Hebert, C. Sminchisescu C., and Y. Weiss. Lecture Notes in Computer Science, vol 11205, pages 587-604. Springer, Cham, 2018. © 2018 Springer Natureen_US
dc.contributor.departmentMIT-IBM Watson AI Lab
dc.contributor.departmentMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
dc.eprint.versionAuthor's final manuscripten_US
dc.type.urihttp://purl.org/eprint/type/ConferencePaperen_US
eprint.statushttp://purl.org/eprint/status/NonPeerRevieweden_US
dc.date.updated2019-07-11T17:31:29Z
dspace.date.submission2019-07-11T17:31:31Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record