DragonPaint: Rule Based Bootstrapping for Small Data with an Application to Cartoon Coloring

K. Gretchen Greene
Proceedings of The 4th International Conference on Predictive Applications and APIs, PMLR 82:1-9, 2018.

Abstract

In this paper, we confront the problem of deep learning’s big labeled data requirements, offer a rule based strategy for extreme augmentation of small data sets and apply that strategy with the image to image translation model by \citetpix2pix:16 to automate cel style cartoon coloring with very limited training data. While our experimental results using geometric rules and transformations demonstrate the performance of our methods on an image translation task with industry applications in art, design and animation, we also propose the use of rules on partial data sets as a generalizable small data strategy, potentially applicable across data types and domains.

Cite this Paper


BibTeX
@InProceedings{pmlr-v82-greene18a, title = {DragonPaint: Rule Based Bootstrapping for Small Data with an Application to Cartoon Coloring}, author = {Greene, K. Gretchen}, booktitle = {Proceedings of The 4th International Conference on Predictive Applications and APIs}, pages = {1--9}, year = {2018}, editor = {Hardgrove, Claire and Dorard, Louis and Thompson, Keiran}, volume = {82}, series = {Proceedings of Machine Learning Research}, month = {24--25 Oct}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v82/greene18a/greene18a.pdf}, url = {https://proceedings.mlr.press/v82/greene18a.html}, abstract = {In this paper, we confront the problem of deep learning’s big labeled data requirements, offer a rule based strategy for extreme augmentation of small data sets and apply that strategy with the image to image translation model by \citetpix2pix:16 to automate cel style cartoon coloring with very limited training data. While our experimental results using geometric rules and transformations demonstrate the performance of our methods on an image translation task with industry applications in art, design and animation, we also propose the use of rules on partial data sets as a generalizable small data strategy, potentially applicable across data types and domains.} }
Endnote
%0 Conference Paper %T DragonPaint: Rule Based Bootstrapping for Small Data with an Application to Cartoon Coloring %A K. Gretchen Greene %B Proceedings of The 4th International Conference on Predictive Applications and APIs %C Proceedings of Machine Learning Research %D 2018 %E Claire Hardgrove %E Louis Dorard %E Keiran Thompson %F pmlr-v82-greene18a %I PMLR %P 1--9 %U https://proceedings.mlr.press/v82/greene18a.html %V 82 %X In this paper, we confront the problem of deep learning’s big labeled data requirements, offer a rule based strategy for extreme augmentation of small data sets and apply that strategy with the image to image translation model by \citetpix2pix:16 to automate cel style cartoon coloring with very limited training data. While our experimental results using geometric rules and transformations demonstrate the performance of our methods on an image translation task with industry applications in art, design and animation, we also propose the use of rules on partial data sets as a generalizable small data strategy, potentially applicable across data types and domains.
APA
Greene, K.G.. (2018). DragonPaint: Rule Based Bootstrapping for Small Data with an Application to Cartoon Coloring. Proceedings of The 4th International Conference on Predictive Applications and APIs, in Proceedings of Machine Learning Research 82:1-9 Available from https://proceedings.mlr.press/v82/greene18a.html.

Related Material