[image 03485] 講演会(4/17)のご案内

Akihiro Sugimoto sugimoto @ nii.ac.jp
2019年 4月 12日 (金) 09:23:19 JST



日時:   4/17(水) 17:00-18:00
場所:   国立情報学研究所 19F 1901号室

講演者: Professor Ming-Hsuan Yang 
              (UC Merced and a Senior Staff Research Scientist at Google)

タイトル: Semantic Segmentation via Domain Adaption and Adversarial Learning

Convolutional neural network-based approaches for semantic
segmentation rely on supervision with pixel-level ground truth, but may not
generalize well to unseen image domains. As the labeling process is tedious
and labor intensive, developing algorithms that can adapt source ground
truth labels to the target domain is of great interest. In the first part
of this talk, I will present an adversarial learning method for domain
adaptation in the context of semantic segmentation. Considering semantic
segmentations as structured outputs that contain spatial similarities
between the source and target domains, we adopt adversarial learning in the
output space. To further enhance the adapted model, we construct a
multi-level adversarial network to effectively perform output space domain
adaptation at different feature levels. Extensive experiments and ablation
study are conducted under various domain adaptation settings, including
synthetic-to-real and cross-city scenarios. We show that the proposed
method performs favorably against the state-of-the-art methods in terms of
accuracy and visual quality. In the second part of this talk, I will
discuss a method for semi-supervised semantic segmentation using an
adversarial network. While most existing discriminators are trained to
classify input images as real or fake on the image level, we design a
discriminator in a fully convolutional manner to differentiate the
predicted probability maps from the ground truth segmentation distribution
with the consideration of the spatial resolution. We show that the proposed
discriminator can be used to improve semantic segmentation accuracy by
coupling the adversarial loss with the standard cross entropy loss of the
proposed model. In addition, the fully convolutional discriminator enables
semi-supervised learning through discovering the trustworthy regions in
predicted results of unlabeled images, thereby providing additional
supervisory signals. In contrast to existing methods that utilize
weakly-labeled images, our method leverages unlabeled images to enhance the
segmentation model. Experimental results on the PASCAL VOC 2012 and
Cityscapes datasets demonstrate the effectiveness of the proposed algorithm.

Bio: Ming-Hsuan Yang <http://faculty.ucmerced.edu/mhyang> a Professor in
Electrical Engineering and Computer Science at University of California at
Merced and a Senior Staff Research Scientist at Google. After receiving his
PhD degree in Computer Science from University of Illinois at
Urbana-Champaign, he worked at Honda Research Institute before joining UC
Merced in 2008. Yang received the Google Faculty Research Award in 2009,
and the Distinguished Early Career Research Award from the UC Merced senate
in 2011, the Faculty Early Career Development (CAREER) award from the
National Science Foundation in 2012, and the Distinguished Research Award
from UC Merced Senate in 2015. He serves as an area chair for several
conferences including IEEE Conference on IEEE International Conference on
Computer Vision (ICCV),  IEEE Computer Vision and Pattern Recognition
(CVPR), European Conference on Computer Vision (ECCV), and Asian Conference
on Computer (ACCV). Yang serves as a program co-chair for ICCV in 2019 as
well as ACCV in 2014, and general co-chair for ACCV in 2016. He serves as
an associate editor of the IEEE Transactions on Pattern Analysis and
Machine Intelligence, International Journal of Computer Vision, Computer
Vision and Image Understanding, Image and Vision Computing, and Journal of
Artificial Intelligence Research. In 2018, he was selected as one of the
Highly Cited Researchers by Clarative Analytics (formerly Thomson Reuters).
Yang received paper awards from UIST 2017, ACCV 2018 and CVPR 2018. He is
an IEEE Fellow.


image メーリングリストの案内