TY - GEN
T1 - Residual Attention Network vs Real Attention on Aesthetic Assessment
AU - Mandal, Ranju
AU - Becken, Susanne
AU - Connolly, Rod M.
AU - Stantic, Bela
N1 - Funding Information:
Acknowledgement. This research was partly funded by the National Environment Science Program (NESP) Tropical Water Quality Hub Project No 5.5.
Publisher Copyright:
© 2021, Springer Nature Singapore Pte Ltd.
PY - 2021
Y1 - 2021
N2 - Photo aesthetics assessment is a challenging problem. Deep Convolutional Neural Network (CNN)-based algorithms have achieved promising results for aesthetics assessment in recent times. Lately, few efficient and effective attention-based CNN architectures are proposed that improve learning efficiency by adaptively adjusts the weight of each patch during the training process. In this paper, we investigate how real human attention affects instead of CNN-based synthetic attention network architecture in image aesthetic assessment. A dataset consists of a large number of images along with eye-tracking information has been developed using an eye-tracking device (https://www.tobii.com/group/about/this-is-eye-tracking/ ) power by sensor technology for our research, and it will be the first study of its kind in image aesthetic assessment. We adopted a Residual Attention Network and ResNet architectures which achieve state-of-the-art performance image recognition tasks on benchmark datasets. We report our findings on photo aesthetics assessment with two sets of datasets consist of original images and images with masked attention patches, which demonstrates higher accuracy when compared to the state-of-the-art methods.
AB - Photo aesthetics assessment is a challenging problem. Deep Convolutional Neural Network (CNN)-based algorithms have achieved promising results for aesthetics assessment in recent times. Lately, few efficient and effective attention-based CNN architectures are proposed that improve learning efficiency by adaptively adjusts the weight of each patch during the training process. In this paper, we investigate how real human attention affects instead of CNN-based synthetic attention network architecture in image aesthetic assessment. A dataset consists of a large number of images along with eye-tracking information has been developed using an eye-tracking device (https://www.tobii.com/group/about/this-is-eye-tracking/ ) power by sensor technology for our research, and it will be the first study of its kind in image aesthetic assessment. We adopted a Residual Attention Network and ResNet architectures which achieve state-of-the-art performance image recognition tasks on benchmark datasets. We report our findings on photo aesthetics assessment with two sets of datasets consist of original images and images with masked attention patches, which demonstrates higher accuracy when compared to the state-of-the-art methods.
KW - Aesthetic scoring
KW - Deep learning
KW - Great Barrier Reef
KW - Image aesthetic evaluation
KW - Photo aesthetic assessment
UR - http://www.scopus.com/inward/record.url?scp=85104734560&partnerID=8YFLogxK
U2 - 10.1007/978-981-16-1685-3_26
DO - 10.1007/978-981-16-1685-3_26
M3 - Conference contribution
AN - SCOPUS:85104734560
SN - 9789811616846
T3 - Communications in Computer and Information Science
SP - 310
EP - 320
BT - Recent Challenges in Intelligent Information and Database Systems - 13th Asian Conference, ACIIDS 2021, Proceedings
A2 - Hong, Tzung-Pei
A2 - Wojtkiewicz, Krystian
A2 - Chawuthai, Rathachai
A2 - Sitek, Pawel
PB - Springer Science and Business Media Deutschland GmbH
T2 - 13th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2021
Y2 - 7 April 2021 through 10 April 2021
ER -