Foreground Object Search by Distilling Composite Image Feature

被引:0
作者
Zhang, Bo [1 ]
Sui, Jiacheng [2 ]
Niu, Li [1 ]
机构
[1] Shanghai Jiao Tong Univ, Artificial Intelligence Inst, Ctr Machine Cognit Comp Artificial Intelligence, Shanghai, Peoples R China
[2] Xi An Jiao Tong Univ, Xian, Peoples R China
来源
2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2023) | 2023年
基金
中国国家自然科学基金;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Foreground object search (FOS) aims to find compatible foreground objects for a given background image, producing realistic composite image. We observe that competitive retrieval performance could be achieved by using a discriminator to predict the compatibility of composite image, but this approach has unaffordable time cost. To this end, we propose a novel FOS method via distilling composite feature (DiscoFOS). Specifically, the abovementioned discriminator serves as teacher network. The student network employs two encoders to extract foreground feature and background feature. Their interaction output is enforced to match the composite image feature from the teacher network. Additionally, previous works did not release their datasets, so we contribute two datasets for FOS task: S-FOSD dataset with synthetic composite images and R-FOSD dataset with real composite images. Extensive experiments on our two datasets demonstrate the superiority of the proposed method over previous approaches.
引用
收藏
页码:22929 / 22938
页数:10
相关论文
共 25 条
[1]  
[Anonymous], 2006, CVPR
[2]  
[Anonymous], 2018, WACV, DOI DOI 10.1109/WACV.2018.00170
[3]  
Chen GB, 2017, ADV NEUR IN, V30
[4]   Improving the Harmony of the Composite Image by Spatial-Separated Attention Module [J].
Cun, Xiaodong ;
Pun, Chi-Man .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2020, 29 :4759-4771
[5]   Deep Parametric Indoor Lighting Estimation [J].
Gardner, Marc-Andre ;
Hold-Geoffroy, Yannick ;
Sunkavalli, Kalyan ;
Gagne, Christian ;
Lalonde, Jean-Francois .
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :7174-7182
[6]  
He KM, 2020, IEEE T PATTERN ANAL, V42, P386, DOI [10.1109/TPAMI.2018.2844175, 10.1109/ICCV.2017.322]
[7]  
Hinton G., 2015, ARXIV
[8]  
Hong Yan, 2022, AAAI
[9]  
Huang Yi-Hua, 2022, CVPR
[10]  
Lalonde JF, 2007, ACM T GRAPHIC, V26, DOI 10.1145/1276377.1276381