Product image retrieval using category-aware siamese convolutional neural network feature
Product image retrieval in the customer-to-shop setting uses similarity learning instead of a predefined distance to address the cross-domain matching problem. Similarity learning can be done using a Siamese convolutional network (SCN) model with pairwise or triplet image sampling. The model trainin...
| Main Authors: | , , |
|---|---|
| Format: | Other |
| Language: | English |
| Published: |
Journal of King Saud University - Computer and Information Sciences
2022
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/284224/1/108.Product%20Image%20Retrieval.pdf |
| _version_ | 1797037815720050688 |
|---|---|
| author | Rahman, Arif Winarko, Edi Mustofa, Khabib |
| author_facet | Rahman, Arif Winarko, Edi Mustofa, Khabib |
| author_sort | Rahman, Arif |
| collection | UGM |
| description | Product image retrieval in the customer-to-shop setting uses similarity learning instead of a predefined distance to address the cross-domain matching problem. Similarity learning can be done using a Siamese convolutional network (SCN) model with pairwise or triplet image sampling. The model training uses product item labels as the target without considering the product category. However, images in the e-shop are inherently have hierarchically structured from the category to the individual image. Therefore, category information should be involved to improve the discriminating factor of the image feature. To accommodate this, we propose a SCN model that involves category and item labels in training to produce the category-aware feature. Our model is based on SCN with modification in training procedure that simultaneously learns the category and item label. Our category-aware Siamese CNN is implemented using MobileNet as the backbone and single-layer network for the mid-feature learner. The results show that our method can improve the accuracy of product image retrieval using SCN based features. © 2022 The Authors |
| first_indexed | 2024-03-14T00:09:44Z |
| format | Other |
| id | oai:generic.eprints.org:284224 |
| institution | Universiti Gadjah Mada |
| language | English |
| last_indexed | 2024-03-14T00:09:44Z |
| publishDate | 2022 |
| publisher | Journal of King Saud University - Computer and Information Sciences |
| record_format | dspace |
| spelling | oai:generic.eprints.org:2842242023-11-30T00:48:05Z https://repository.ugm.ac.id/284224/ Product image retrieval using category-aware siamese convolutional neural network feature Rahman, Arif Winarko, Edi Mustofa, Khabib Artificial Intelligence and Image Processing not elsewhere classified Product image retrieval in the customer-to-shop setting uses similarity learning instead of a predefined distance to address the cross-domain matching problem. Similarity learning can be done using a Siamese convolutional network (SCN) model with pairwise or triplet image sampling. The model training uses product item labels as the target without considering the product category. However, images in the e-shop are inherently have hierarchically structured from the category to the individual image. Therefore, category information should be involved to improve the discriminating factor of the image feature. To accommodate this, we propose a SCN model that involves category and item labels in training to produce the category-aware feature. Our model is based on SCN with modification in training procedure that simultaneously learns the category and item label. Our category-aware Siamese CNN is implemented using MobileNet as the backbone and single-layer network for the mid-feature learner. The results show that our method can improve the accuracy of product image retrieval using SCN based features. © 2022 The Authors Journal of King Saud University - Computer and Information Sciences 2022 Other NonPeerReviewed application/pdf en https://repository.ugm.ac.id/284224/1/108.Product%20Image%20Retrieval.pdf Rahman, Arif and Winarko, Edi and Mustofa, Khabib (2022) Product image retrieval using category-aware siamese convolutional neural network feature. Journal of King Saud University - Computer and Information Sciences. https://www.sciencedirect.com/science/article/pii/S1319157822000763?via%3Dihub 10.1016/j.jksuci.2022.03.005 |
| spellingShingle | Artificial Intelligence and Image Processing not elsewhere classified Rahman, Arif Winarko, Edi Mustofa, Khabib Product image retrieval using category-aware siamese convolutional neural network feature |
| title | Product image retrieval using category-aware siamese convolutional neural network feature |
| title_full | Product image retrieval using category-aware siamese convolutional neural network feature |
| title_fullStr | Product image retrieval using category-aware siamese convolutional neural network feature |
| title_full_unstemmed | Product image retrieval using category-aware siamese convolutional neural network feature |
| title_short | Product image retrieval using category-aware siamese convolutional neural network feature |
| title_sort | product image retrieval using category aware siamese convolutional neural network feature |
| topic | Artificial Intelligence and Image Processing not elsewhere classified |
| url | https://repository.ugm.ac.id/284224/1/108.Product%20Image%20Retrieval.pdf |
| work_keys_str_mv | AT rahmanarif productimageretrievalusingcategoryawaresiameseconvolutionalneuralnetworkfeature AT winarkoedi productimageretrievalusingcategoryawaresiameseconvolutionalneuralnetworkfeature AT mustofakhabib productimageretrievalusingcategoryawaresiameseconvolutionalneuralnetworkfeature |