Learning to assess visual aesthetics of food images

doi:10.1007/s41095-020-0193-5

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 多媒体计算与图形学团队

	Learning to assess visual aesthetics of food images
	Sheng, Kekai 2,3; Dong, Weiming 3; Huang, Haibin 5; Chai, Menglei 1; Zhang, Yong 4; Ma, Chongyang 5; Hu, Bao-Gang 3
刊名	COMPUTATIONAL VISUAL MEDIA
	2021-03-01
卷号	7 期号:1 页码:139-152
关键词	image aesthetic assessment food image analysis dataset regularization
ISSN号	2096-0433
DOI	10.1007/s41095-020-0193-5
通讯作者	Dong, Weiming(weiming.dong@ia.ac.cn)
英文摘要	Distinguishing aesthetically pleasing food photos from others is an important visual analysis task for social media and ranking systems related to food. Nevertheless, aesthetic assessment of food images remains a challenging and relatively unexplored task, largely due to the lack of related food image datasets and practical knowledge. Thus, we present the Gourmet Photography Dataset (GPD), the first large-scale dataset for aesthetic assessment of food photos. It contains 24,000 images with corresponding binary aesthetic labels, covering a large variety of foods and scenes. We also provide a non-stationary regularization method to combat over-fitting and enhance the ability of tuned models to generalize. Quantitative results from extensive experiments, including a generalization ability test, verify that neural networks trained on the GPD achieve comparable performance to human experts on the task of aesthetic assessment. We reveal several valuable findings to support further research and applications related to visual aesthetic analysis of food images. To encourage further research, we have made the GPD publicly available at https://github.com/Openning07/GPA.
资助项目	National Natural Science Foundation of China[61832016] ; National Natural Science Foundation of China[61672520] ; CASIA-Tencent Youtu joint research project
WOS关键词	NETWORKS
WOS研究方向	Computer Science
语种	英语
出版者	SPRINGERNATURE
WOS记录号	WOS:000648692800009
资助机构	National Natural Science Foundation of China ; CASIA-Tencent Youtu joint research project
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/44633]
专题	自动化研究所_模式识别国家重点实验室_多媒体计算与图形学团队
通讯作者	Dong, Weiming
作者单位	1.Snap Inc, Santa Monica, CA 90405 USA 2.Tencent, Youtu Lab, Shanghai 200233, Peoples R China 3.Chinese Acad Sci, Inst Automat, NLPR, Beijing 100190, Peoples R China 4.Tencent Inc, AI Lab, Shenzhen 518000, Peoples R China 5.Kuaishou Technol, Beijing 100085, Peoples R China
推荐引用方式 GB/T 7714	Sheng, Kekai,Dong, Weiming,Huang, Haibin,et al. Learning to assess visual aesthetics of food images[J]. COMPUTATIONAL VISUAL MEDIA,2021,7(1):139-152.
APA	Sheng, Kekai.,Dong, Weiming.,Huang, Haibin.,Chai, Menglei.,Zhang, Yong.,...&Hu, Bao-Gang.(2021).Learning to assess visual aesthetics of food images.COMPUTATIONAL VISUAL MEDIA,7(1),139-152.
MLA	Sheng, Kekai,et al."Learning to assess visual aesthetics of food images".COMPUTATIONAL VISUAL MEDIA 7.1(2021):139-152.