Articulation of features of fashion into keywords is an arduous task. Description of fashion seen on other people is insufficient, not to mention, inaccurate for a conventional search engine that takes keywords as queries. To overcome this shortcoming, this paper outlines a model for a search engine that takes a fashion image as a query and returns five most similar images from its database. The model consists of CNN classification model that classifies the query image into one of the five classes and a Convolutional Autoencoder that returns five images with most similar features from that class of images. Similarity between images is found by calculating the similarity of the encoded vector of the query image with the encoded vectors of the images in the class predicted by the classifier. Since the encoding and decoding is done by the Autoencoder based on the nature of the images given for training, the model returns images based on features that prove important enough to be encoded based on the training images. In other words, the features that are necessary to ensure as little loss as possible when decoding the encoded vector. This forms the basis for using similarity between encoded vectors to find similar images to the given query image. The model is trained using fashion images to find similar fashion to query image.

Classification, Similar images, CNN, Autoencoder, Clothes

[1] S. Ota, H. Takenouchi and M. Tokumaru, “Kansei clothing retrieval system using features extracted by autoencoder”, 1-7. 10.1109/SSCI.2017.8285320 (2017).
[2] Z.-Q. Cheng, X. Wu, Y. Liu, and X.-S. Hua. “Video2shop: Exact matching clothes in videos to online shopping images”, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 4048–4056, (2017)
[3] A. Y. Ivanov, G. I. Borzunov and K. Kogos, "Recognition and identification of the clothes in the photo or video using neural networks", 2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus), Moscow, 2018, pp. 1513-1516.
[4] X.Wang, Z.Sun,W.Zhang, Y.Zhou,and Y.Jiang. “Matching user photos to online products with robust deep features”, In ICMR, pages 7–14, (2016).
[5] L. Ami D, M. Shilpa I, R. Urvashi K., S. Saliha H. “Reverse Image Search Image retrieval using various techniques”, International Journal of Advance Engineering and Research Development (IJAERD) Volume 4, Issue 4, April-2017, e-ISSN: 2348 - 4470, print-ISSN: 2348-6406.
[6] M. M. Singh, C. David, P. M. Rathod, "Smart Image Search Android application using Reverse Image Search", IJREAM, vol. 2, no. 1, pp. 1-4, (2016).
[7] Kaur, M., & Sohi, N. “A novel technique for content-based image retrieval using colour, texture and edge features”, In Proceedings of the international conference on communication and electronics systems (ICCES), (2016)
[8] Huang, C., Wei, C., Wang, Y. “Active learning-based clothing image recommendation with implicit user preferences”, In ICME Workshops, pp. 1–4 (2013)
[9] J. Rani, S & Valli Kumari, V. “Improved Hill Climbing Based Segmentation (IHCBS) technique for CBIR system”, 1-10. 10.1109/ICECCT.2015.7226069 (2015).
[10] K. Yamaguchi, M. H. Kiapour, and T. L. Berg. “Paper doll parsing: Retrieving similar styles to parse clothing items”. In ICCV, (2013).
[11] Adrian Rosebrock, “How to (quickly) build a deep learning image dataset”, pyimagesearch.com, https://www.pyimagesearch.com/2018/04/09/how-to-quickly-build-a-deep-learning-image-dataset/ (accessed Apr.27, 2019)
[12] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna. “Rethinking the inception architecture for computer vision”, In CVPR, 2016.
[13] M. Yang and K. Yu. “Real-time clothing recognition in surveillance videos”, In ICIP, (2011).
[14] I. Goodfellow, Y. Bengio, A. Courville, “Deep Learning”, Cambridge, MA, USA, MIT Press, ch. 14 ,2016.
[15] V. Jayawardana, “Autoencoders — Bits and Bytes of Deep Learning”, towardsdatascience.com, https://towardsdatascience.com/autoencoders-bits-and-bytes-of-deep-learning-eaba376f23ad (accessed Apr. 27, 2019).
[16] The SciPy community, “numpy.linalg.norm”, docs.scipy.org, https://docs.scipy.org/doc/numpy-1.14.0/reference/generated/numpy.linalg.norm.html#numpy-linalg-norm (accessed Apr. 27, 2019).
[17] M. H. Kiapour, X. Han, S. Lazebnik, A. C. Berg, and T. L. Berg, “Where to buy it: Matching street clothing photos in online shops”, In ICCV, (2015)
[18] A. Y. Ivanov, G. I. Borzunov and K. Kogos, "Recognition and identification of the clothes in the photo or video using neural networks," 2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus), Moscow, 2018, pp. 1513-1516.
[19] Z. Liu, P. Luo, S. Qiu, X. Wang, and X. Tang. “Deepfashion: Powering robust clothes recognition and retrieval with rich annotations”, In CVPR, pages 1096–1104, 2016.
[20] S. Thompson, M. Reilly, "A picture is worth a thousand words: Reverse image lookup and digital library assessment", J. Assoc. Inf. Sci. Technol., vol. 68, pp. 2264-2266, 2017, DOI: 10.1002/asi.
[21] P. Nieuwenhuysen, "Search by image through the Internet: applications and limitations", Libraries in the Transition Era: New Space –New Services – New Experience: Proc. of the 7th Shanghai International Library Forum, pp. 145-155, (2014), [online] Available: http://www.vub.ac.be/BIBLIO/nieuwenhuysen/presentations/.
[22] Punam Mahesh Ingale, "The importance of Digital Image Processing and its applications", International Journal of Scientific Research in Computer Science and Engineering, Vol.06, Issue.01, pp.31-32, 2018.
[23] N.S. Lele , "Image Classification Using Convolutional Neural Network", International Journal of Scientific Research in Computer Science and Engineering, Vol.6, Issue.3, pp.22-26, 2018.
[24] John Bosco.P., S.K.V. Jayakumar, "A Study on Web Based Image Search by Re-Ranking Techniques", International Journal of Scientific Research in Network Security and Communication, Vol.5, Issue.3, pp.19-26, 2017.
[25] Mahajan J.R., C.S. Rawat, "Object Detection and Tracking using Congnitive Approach", International Journal of Scientific Research in Network Security and Communication, Vol.5, Issue.3, pp.136-140, 2017.