LPaintB: Learning to Paint from Self-Supervision

dc.contributor.authorJia, Biaoen_US
dc.contributor.authorBrandt, Jonathanen_US
dc.contributor.authorMech, Radomíren_US
dc.contributor.authorKim, Byungmoonen_US
dc.contributor.authorManocha, Dineshen_US
dc.contributor.editorLee, Jehee and Theobalt, Christian and Wetzstein, Gordonen_US
dc.date.accessioned2019-10-14T05:12:44Z
dc.date.available2019-10-14T05:12:44Z
dc.date.issued2019
dc.description.abstractWe present a novel reinforcement learning-based natural media painting algorithm. Our goal is to reproduce a reference image using brush strokes and we encode the objective through observations. Our formulation takes into account that the distribution of the reward in the action space is sparse and training a reinforcement learning algorithm from scratch can be difficult. We present an approach that combines self-supervised learning and reinforcement learning to effectively transfer negative samples into positive ones and change the reward distribution. We demonstrate the benefits of our painting agent to reproduce reference images with brush strokes. The training phase takes about one hour and the runtime algorithm takes about 30 seconds on a GTX1080 GPU reproducing a 1000x800 image with 20,000 strokes.en_US
dc.description.sectionheadersImages and Learning
dc.description.seriesinformationPacific Graphics Short Papers
dc.identifier.doi10.2312/pg.20191336
dc.identifier.isbn978-3-03868-099-4
dc.identifier.issn-
dc.identifier.pages33-39
dc.identifier.urihttps://doi.org/10.2312/pg.20191336
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/pg20191336
dc.publisherThe Eurographics Associationen_US
dc.titleLPaintB: Learning to Paint from Self-Supervisionen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
033-039.pdf
Size:
2.19 MB
Format:
Adobe Portable Document Format