Augmented reality is an emerging technology in many application domains.
Among them is the beauty industry, where live virtual try-on of beauty products
is of great importance. In this paper, we address the problem of live hair
color augmentation. To achieve this goal, hair needs to be segmented quickly
and accurately. We show how a modified MobileNet CNN architecture can be used
to segment the hair in real-time. Instead of training this network using large
amounts of accurate segmentation data, which is difficult to obtain, we use
crowd sourced hair segmentation data. While such data is much simpler to
obtain, the segmentations there are noisy and coarse. Despite this, we show how
our system can produce accurate and fine-detailed hair mattes, while running at
over 30 fps on an iPad Pro tablet.
%0 Generic
%1 citeulike:14514542
%A xxx,
%D 2017
%K loss segmentation upsampling
%T Real-time deep hair matting on mobile devices
%U http://arxiv.org/abs/1712.07168
%X Augmented reality is an emerging technology in many application domains.
Among them is the beauty industry, where live virtual try-on of beauty products
is of great importance. In this paper, we address the problem of live hair
color augmentation. To achieve this goal, hair needs to be segmented quickly
and accurately. We show how a modified MobileNet CNN architecture can be used
to segment the hair in real-time. Instead of training this network using large
amounts of accurate segmentation data, which is difficult to obtain, we use
crowd sourced hair segmentation data. While such data is much simpler to
obtain, the segmentations there are noisy and coarse. Despite this, we show how
our system can produce accurate and fine-detailed hair mattes, while running at
over 30 fps on an iPad Pro tablet.
@misc{citeulike:14514542,
abstract = {{Augmented reality is an emerging technology in many application domains.
Among them is the beauty industry, where live virtual try-on of beauty products
is of great importance. In this paper, we address the problem of live hair
color augmentation. To achieve this goal, hair needs to be segmented quickly
and accurately. We show how a modified MobileNet CNN architecture can be used
to segment the hair in real-time. Instead of training this network using large
amounts of accurate segmentation data, which is difficult to obtain, we use
crowd sourced hair segmentation data. While such data is much simpler to
obtain, the segmentations there are noisy and coarse. Despite this, we show how
our system can produce accurate and fine-detailed hair mattes, while running at
over 30 fps on an iPad Pro tablet.}},
added-at = {2019-02-27T22:23:29.000+0100},
archiveprefix = {arXiv},
author = {xxx},
biburl = {https://www.bibsonomy.org/bibtex/26bf7e9f4c17e4d64119aa8e5536725c9/nmatsuk},
citeulike-article-id = {14514542},
citeulike-linkout-0 = {http://arxiv.org/abs/1712.07168},
citeulike-linkout-1 = {http://arxiv.org/pdf/1712.07168},
day = 19,
eprint = {1712.07168},
interhash = {9c0bf3d52d56fbc7120e635e29d962ea},
intrahash = {6bf7e9f4c17e4d64119aa8e5536725c9},
keywords = {loss segmentation upsampling},
month = dec,
posted-at = {2018-01-10 14:35:24},
priority = {3},
timestamp = {2019-02-27T22:23:29.000+0100},
title = {{Real-time deep hair matting on mobile devices}},
url = {http://arxiv.org/abs/1712.07168},
year = 2017
}