TY - GEN
T1 - Shadow detection and removal using GAN
AU - Nagae, Takahiro
AU - Abiko, Ryo
AU - Yamaguchi, Takuro
AU - Ikehara, Masaaki
N1 - Publisher Copyright:
© 2021 European Signal Processing Conference, EUSIPCO. All rights reserved.
PY - 2021/1/24
Y1 - 2021/1/24
N2 - To remove shadowed region in a single image, it is important to obtain high accuracy in both two processes, shadow detection and removal. In order to improve the results, recent methods perform these two processes simultaneously and use GAN for the training. However, since these methods do not try to maintain the luminance of non-shadowed regions, the output images tend to be faded. In this paper, to overcome fading problem, we proposed a new GAN structure based on shadow model. Since our GAN-based method focus on the variation of the illuminance, the illuminances of the shadowed regions, whose amount of change are large, are effectively estimated. In addition, non-shadowed regions remain slightly faded due to our new GAN structure and training method. Owing to our novel GAN structure and training method, our method outperforms state-of-the-art methods in PSNR and SSIM.
AB - To remove shadowed region in a single image, it is important to obtain high accuracy in both two processes, shadow detection and removal. In order to improve the results, recent methods perform these two processes simultaneously and use GAN for the training. However, since these methods do not try to maintain the luminance of non-shadowed regions, the output images tend to be faded. In this paper, to overcome fading problem, we proposed a new GAN structure based on shadow model. Since our GAN-based method focus on the variation of the illuminance, the illuminances of the shadowed regions, whose amount of change are large, are effectively estimated. In addition, non-shadowed regions remain slightly faded due to our new GAN structure and training method. Owing to our novel GAN structure and training method, our method outperforms state-of-the-art methods in PSNR and SSIM.
KW - GAN
KW - Illuminance
KW - Shadow detection
KW - Shadow removal
UR - http://www.scopus.com/inward/record.url?scp=85099276158&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85099276158&partnerID=8YFLogxK
U2 - 10.23919/Eusipco47968.2020.9287528
DO - 10.23919/Eusipco47968.2020.9287528
M3 - Conference contribution
AN - SCOPUS:85099276158
T3 - European Signal Processing Conference
SP - 630
EP - 634
BT - 28th European Signal Processing Conference, EUSIPCO 2020 - Proceedings
PB - European Signal Processing Conference, EUSIPCO
T2 - 28th European Signal Processing Conference, EUSIPCO 2020
Y2 - 24 August 2020 through 28 August 2020
ER -