Dominant Color Detection For Online Fashion Retrievals

Yıl 2024, Cilt: 14 Sayı: 1, 69 - 80, 07.07.2024

Sultan Zeybek , Merve Çelik

https://doi.org/10.55024/buyasambid.1501329

Öz

This paper introduces a novel approach aimed at efficiently extracting dominant colors from online fashion images. The method addresses challenges related to detecting overlapping objects and computationally expensive methods by combining K-means clustering and graph-cut techniques into a framework. This framework incorporates an adaptive weighting strategy to enhance color extraction accuracy. Additionally, it introduces a two-phase fashion apparel detection method called YOLOv4, which utilizes U-Net architecture for clothing segmentation to precisely separate clothing items from the background or other elements. Experimental results show that K-means with YOLOv4 outperforms K-means with the U-Net model. These findings suggest that the U-Net architecture and YOLOv4 models can be effective methods for complex image segmentation tasks in online fashion retrieval and image processing, particularly in the rapidly evolving e-commerce environment.

Anahtar Kelimeler

Dominant color extraction, Fashion image analysis, K-means clustering, Image Segmentation

Çevrimiçi Moda Aramaları İçin Baskın Renk Tespiti

Öz

Bu makale, çevrimiçi moda görüntülerinden baskın renklerin verimli bir şekilde çıkarılmasını amaçlayan yeni bir yaklaşımı tanıtmaktadır. Bu yöntem, üst üste binen nesnelerin tespit edilmesi sırasında ortaya çıkan zorluklara ve hesaplama maliyeti yüksek yöntemlere çözüm sunarak, K-means kümeleme ve graf-kesim tekniklerini birleştiren ve adaptif bir ağırlıklandırma stratejisi kullanılarak renk çıkarımının doğruluğu artırmayı amaçlayan bir çerçeve üzerine kurulmuştur. Giysi segmentasyonu için U-Net mimarisini ile giysi öğelerini arka plandan veya diğer unsurlardan hassas bir şekilde ayırmayı sağlayarak giysi öznitelik tahmini ve ayrıştırma görevi için YOLOv4 adlı iki aşamalı bir moda giyim tespit yöntemini tanıtmaktadır ile karşılaştırılmıştır. Deneysel sonuçlar, K-means ile YOLOv4'ün, K-means ile U-Net modeline kıyasla daha üstün performans sergilediğini göstermektedir. Bu bulgular, özellikle hızla gelişen e-ticaret ortamında çevrimiçi moda arama ve görüntü işleme alanlarında ilerlemeye katkı sağlamak amacıyla U-Net mimarisinin ve YOLOv4 mimarilerinin karmaşık görüntü segmentasyon görevlerini için etkin metotlar olarak kullanılabileceğini göstermiştir.

Anahtar Kelimeler

Moda Görüntü Analizi, Baskın Renk Tespiti, K-means Kümeleme, Görüntü Bölütleme

Kaynakça

• Agrawal, S., Panda, R., Choudhury, P., & Abraham, A. (2022). Dominant color component and adaptive whale optimization algorithm for multilevel thresholding of color images. Knowledge-Based Systems, 240, 108172. https://doi.org/10.1016/j.knosys.2021.108172.
• Agrawal, S., Panda, R., Choudhury, P., & Abraham, A. (2022). Dominant color component and adaptive whale optimization algorithm for multilevel thresholding of color images. Knowledge-Based Systems, 240, 108172.
• Bu, Q., Zeng, K., Wang, R., & Feng, J. (2020). Multi-depth dilated network for fashion landmark detection with batch-level online hard keypoint mining. Image and Vision Computing, 99, 103930. https://doi.org/10.1016/j.imavis.2019.103930
• Chang, Y., & Mukai, N. (2022). Color feature based dominant color extraction. IEEE Access, 10, 93055-93061. https://doi.org/10.1109/ACCESS.2022.3202632.
• Chang, Y., Iida, T., and Mukai, N. (2015). Dominant color extraction method from natural images. Proceedings of the International Conference on Image Processing, 44, 637-643.
• Gunduz, A. B., Taskin, B., Yavuz, A. G., & Karsligil, M. E. (2021). A better way of extracting dominant colors using salient objects with semantic segmentation. Engineering Applications of Artificial Intelligence, 100, 104204. https://doi.org/10.1016/j.engappai.2021.104204.
• Gunduz, A. B., Taskin, B., Yavuz, A. G., & Karsligil, M. E. (2021). A better way of extracting dominant colors using salient objects with semantic segmentation. Engineering Applications of Artificial Intelligence, 100, 104204.
• Hu, Z., Yan, H., & Lin, X. (2008). Clothing segmentation using foreground and background estimation based on the constrained Delaunay triangulation. Pattern Recognition, 41(5), (pp. 1581-1592)
• Ilea, D. E., & Whelan, P. F. (2011). Image segmentation based on the integration of colour–texture descriptors—A review. Pattern Recognition, 44(10–11), 2479-2501. https://doi.org/10.1016/j.patcog.2011.03.011
• Ilea, D., & Whelan, P. (2006). Color image segmentation using a spatial k-means clustering algorithm. Proceedings of the 10th International Machine Vision and Image Processing Conference, 30-1.
• Kalantidis, Y., Kennedy, L., & Li, L.-J. (2013). Getting the look: Clothing recognition and segmentation for automatic product suggestions in everyday photos. In Proceedings of the 3rd ACM Conference on International Conference on Multimedia Retrieval, ICMR, (pp. 105–112) Association for Computing Machinery, New York, NY, USA.
• Karthick, P., Mohiuddine, S. A., Tamilvanan, K., Narayanamoorthy, S., & Maheswari, S. (2023). Investigations of color image segmentation based on connectivity measure, shape priority and normalized fuzzy graph cut. Applied Soft Computing, 139, 110239.
• Khotanzad, A., & Hernandez, O. J. (2003). Color image retrieval using multispectral random field texture model and color content features. Pattern Recognition, 36(8), 1679-1694. https://doi.org/10.1016/S0031-3203(02)00320-4
• Khotanzad, A., & Hernandez, O. J. (2003). Color image retrieval using multispectral random field texture model and color content features. Pattern Recognition, 36(8), 1679-1694.
• Lai, P., & Westland, S. (2020). Machine learning for colour palette extraction from fashion runway images. International Journal of Fashion Design, Technology and Education, 13(3), 334-340.
• Lee, C. H., & Lin, C. W. (2021). A two-phase fashion apparel detection method based on YOLOv4. Applied Sciences, 11(9), 3782.
• Lee, C.-H., & Lin, C.-W. (2021). A two-phase fashion apparel detection method based on YOLOv4. Applied Sciences, 11(9), 3782. https://doi.org/10.3390/app11093782
• Liang, X., Lin, L., Yang, W., Luo, P., Huang, J., & Yan, S. (2016). Clothes Co-Parsing Via Joint Image Segmentation and Labeling With Application to Clothing Retrieval. IEEE Transactions on Multimedia, 18(6), https://doi.org/10.1109/TMM.2016.2553482
• Liao, L., He, X., Zhao, B., Ngo, C.-W., & Chua, T.-S. (2018). Interpretable Multimodal Retrieval for Fashion Products. In Proceedings of the 26th ACM International Conference on Multimedia (MM '18) (pp. 1571–1579). Association for Computing Machinery
• Lin, C.-H., Chen, R.-T., & Chan, Y., K. (2009). A smart content-based image retrieval system based on color and texture features. Image and Vision Computing, 27(6), 658-665. https://doi.org/10.1016/j.imavis.2008.07.002
• Liu, Y., Wan, Z., Yin, X., Yue, G., Tan, A., & Zheng, Z. (2023). Detection of GAN generated image using color gradient representation. Journal of Visual Communication and Image Representation, 95, 103876. https://doi.org/10.1016/j.jvcir.2022.103876.
• Liu, Z., Luo, P., Qiu, S., Wang, X., & Tang, X. (2016). DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations. I2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1096-1104.
• Liu, Z.-y., Ding, F., Xu, Y., & Han, X. (2021). Background dominant colors extraction method based on color image quick fuzzy c-means clustering algorithm. Defence Technology, 17*(5), (pp1782–1790. https://doi.org/10.1016/j.dt.2021.03.020.
• Liu,H.,Wang,Y.,Chen,D.,Lv,J. & Alshalabi,R.(2023).Garment Image Retrieval based on Grab Cut Auto Segmentation and Dominate Color Method. Applied Mathematics and Nonlinear Sciences,8(1) 573-584. https://doi.org/10.2478/amns.2022.2.0042
• Lu, B., Zhou, J., Wang, Q., Zou, G., & Yang, J. (2023). Fusion-based color and depth image segmentation method for rocks on conveyor belt. Minerals Engineering, 199, 108107. https://doi.org/10.1016/j.mineng.2023.108107.
• Lu, H., Gao, Q., Zhang, X., & Xia, W. (2022). A multi-view clustering framework via integrating K-means and graph-cut. Neurocomputing, 501, 609-617. https://doi.org/10.1016/j.neucom.2022.02.082.
• Lu, Y., Young, S., Wang, H., & Wijewardane, N. (2022). Robust plant segmentation of color images based on image contrast optimization. Computers and Electronics in Agriculture, 193, 106711. https://doi.org/10.1016/j.compag.2022.106711.
• Mezaris, V., & Kompatsiaris, I. (2004). Real-Time Compressed-Domain Spatiotemporal Segmentation and Ontologies for Video Indexing and Retrieval. IEEE Transactions on Circuit and Systems for Video Technology, 14(5). https://doi.org/10.1109/TCSVT.2004.828272
• Ngọc, M. Ô. V., Carlinet, E., Fabrizio, J., & Géraud, T. (2023). The Dahu graph cut for interactive segmentation on 2D/3D images. Pattern Recognition, 136, 109207. https://doi.org/10.1016/j.patcog.2021.109207.
• P Karthick, P., Mohiuddine, S. A., Tamilvanan, K., Narayanamoorthy, S., & Maheswari, S. (2023). Investigations of color image segmentation based on connectivity measure, shape priority and normalized fuzzy graph cut. Applied Soft Computing, 139, 110239. https://doi.org/10.1016/j.asoc.2023.110239.
• Park, S., Shin, M., Ham, S., Choe, S., & Kang, Y. (2019). Study on fashion image retrieval methods for efficient fashion visual search. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA, (pp. 316-319), doi: 10.1109/CVPRW.2019.00042.
• S. Park, M. Shin, S. Ham, S. Choe & Y. Kang, (2019). Study on fashion image retrieval methods for efficient fashion visual search. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, (pp. 316-319), Long Beach, CA, USA.
• Saranya, M. S., & Geetha, P. (2023). Cross-domain fashion cloth retrieval via novel attention-guided cascade neural network and clothing parsing. Computer Vision and Image Understanding, 235, 103777. https://doi.org/10.1016/j.cviu.2021.104204
• Shih, H.-C., & Liu, E.-R. (2016). New quartile-based region merging algorithm for unsupervised image segmentation using color-alone feature. Information Sciences, 342, 24–36. https://doi.org/10.1016/j.ins.2016.01.029.
• Talib, A., Mahmuddin, M., Husni, H., & George, L. E. (2013). A weighted dominant color descriptor for content-based image retrieval. Journal of Visual Communication and Image Representation, 24(3), 345–360. https://doi.org/10.1016/j.jvcir.2012.12.008.
• Tomasi, C., & Manduchi, R. (1998). Bilateral filtering for gray and color images. In Sixth International Conference on Computer Vision, pp. 839–846. IEEE.
• Wang, X. (2019). Towards color compatibility in fashion using machine learning.
• Wang, X. Y., Zhang, X. J., Yang, H. Y., & Bu, J. (2012). A pixel-based color image segmentation using support vector machine and fuzzy C-means. Neural Networks, 33, 148-159.
• Wang, X.-Y., Zhang, X.-J., Yang, H.-Y., & Bu, J. (2012). A pixel-based color image segmentation using support vector machine and fuzzy C-means. Neural Networks, 33, 148–159. https://doi.org/10.1016/j.neunet.2012.03.018. Yamaguchi, K., Kiapour, M. H., Ortiz, L. E., & Berg, T. L. (2012). Parsing clothing in fashion photographs. In CVPR.
• Yamaguchi, K., Kiapour, M. H., Ortiz, L. E., & Berg, T. L. (2012). Parsing clothing in fashion photographs. In 2012 IEEE Conference on Computer vision and pattern recognition (pp. 3570-3577). IEEE.
• Yu, L. L., Simo-Serra, E., Moreno-Noguer, F., & Rubio, A. (2017). Multi-modal Embedding for Main Product Detection in Fashion. In Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017 (pp. 2236-2242). Institute of Electrical and Electronics Engineers Inc. https://doi.org/10.1109/ICCVW.2017.261