DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yoo, Seungjoo | ko |
dc.contributor.author | Bahng, Hyojin | ko |
dc.contributor.author | Chung, Sunghyo | ko |
dc.contributor.author | Lee, Junsoo | ko |
dc.contributor.author | Chang, Jaehyuk | ko |
dc.contributor.author | Choo, Jaegul | ko |
dc.date.accessioned | 2021-01-12T02:50:20Z | - |
dc.date.available | 2021-01-12T02:50:20Z | - |
dc.date.created | 2020-12-03 | - |
dc.date.issued | 2019-06 | - |
dc.identifier.citation | 32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2019, pp.11275 - 11284 | - |
dc.identifier.uri | http://hdl.handle.net/10203/279876 | - |
dc.description.abstract | Despite recent advancements in deep learning-based automatic colorization, they are still limited when it comes to few-shot learning. Existing models require a significant amount of training data. To tackle this issue, we present a novel memory-augmented colorization model MemoPainter that can produce high-quality colorization with limited data. In particular, our model is able to capture rare instances and successfully colorize them. Also, we propose a novel threshold triplet loss that enables unsupervised training of memory networks without the need for class labels. Experiments show that our model has superior quality in both few-shot and one-shot colorization tasks. | - |
dc.language | English | - |
dc.publisher | IEEE Computer Society | - |
dc.title | Coloring with limited data: Few-shot colorization via memory augmented networks | - |
dc.type | Conference | - |
dc.identifier.wosid | 000542649304090 | - |
dc.identifier.scopusid | 2-s2.0-85078777565 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 11275 | - |
dc.citation.endingpage | 11284 | - |
dc.citation.publicationname | 32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2019 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Long Beach | - |
dc.identifier.doi | 10.1109/CVPR.2019.01154 | - |
dc.contributor.localauthor | Choo, Jaegul | - |
dc.contributor.nonIdAuthor | Yoo, Seungjoo | - |
dc.contributor.nonIdAuthor | Bahng, Hyojin | - |
dc.contributor.nonIdAuthor | Chung, Sunghyo | - |
dc.contributor.nonIdAuthor | Lee, Junsoo | - |
dc.contributor.nonIdAuthor | Chang, Jaehyuk | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.