Non-personalized Fashion Outfit Recommendations

The Problem of Cold Starts
  • Anna Iliukovich-Strakovskaia
  • Victoria Tsvetkova
  • Emeli Dral
  • Alexey Dral
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 747)

Abstract

Nowadays the demand for compatible outfit recommendation system is rising up. There are a number of online fashion shops and web applications allowing their customers to observe collages of fashionable items which look good together. Such forms of recommendations help both clients and sellers. On one hand clients can make better choices while shopping and find new items of clothing that will match each other. On the other hand, complex outfit recommendations help sellers to sell more items which will influence their business KPIs (key performance indicators).

We can build fashion outfit recommendation system using various instruments: machine learning (mostly supervised learning), statistics, user modelling, rules construction and many others. The problem is that most of these techniques require a solid amount of data about users’ preferences and tastes to create good personal recommendations. But what if there is no such information? In this paper we are proposing the approach of creating fashion outfit recommendations in situations when we lack of that data (the cold start problem). We are going to share (1) the approach of dataset gathering and (2) the results of using external datasets and pre-trained models for fashion outfit recommendation creation.

Keywords

Fashion outfit recommendations Recommender systems Machine learning Neural networks Transfer learning 

Notes

Acknowledgments

This work has been completed at the chair of Algorithms and Technologies of Programming, Department of Innovation and High Technology, Moscow Institute of Physics and Technology (ATP DIHT MIPT).

References

  1. 1.
  2. 2.
    Chen, Q., Wang, G., Tan, C.L.: Modeling fashion. In: 2013 IEEE International Conference on Multimedia and Expo (ICME). IEEE (2013)Google Scholar
  3. 3.
    Hu, Y., Koren, Y., Volinsky, C.: Collaborative filtering for implicit feedback datasets. In: Eighth IEEE International Conference on Data Mining, 2008, ICDM 2008. IEEE (2008)Google Scholar
  4. 4.
    Iliukovich-Strakovskaia, A., Dral, A., Dral, E.: Using pre-trained models for fine-grained image classification in fashion field. In: KDD 2016 Workshop on Machine Learning Meets Fashion (2016)Google Scholar
  5. 5.
    Tangseng, P., Yamaguchi, K., Okatani, T.: Recommending outfits from personal closet. In: The IEEE International Conference on Computer Vision (ICCV) (2017)Google Scholar
  6. 6.
    Zhu, S., Fidler, S., Urtasun, R., Lin, D., Loy, C.C.: Be your own prada: fashion synthesis with structural coherence. arXiv:1710.07346 (2017)
  7. 7.
    Szegedy, C., et al.: Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2015)Google Scholar
  8. 8.
    Chen, T., Li, M., Xiao, T., Li, Y., Xu, B., Lin, M., Wang, N., Wang, M., Zhang, C., Zhang, Z.: MXNet: a flexible and efficient machine learning library for heterogeneous distributed systems (2015)Google Scholar
  9. 9.
    Kang, W.-C., Fang, C., Wang, Z., McAuley, J.: Visually-aware fashion recommendation and design with generative image models. arXiv:1711.02231 (2017)
  10. 10.
    Wang, J., et al.: Towards predicting the likeability of fashion images. arXiv preprint arXiv:1511.05296 (2015)
  11. 11.
    Han, X., Wu, Z., Jiang, Y.-G., Davis, L.S.: Learning fashion compatibility with bidirectional LSTMs. arXiv:1707.05691 (2017)
  12. 12.
    Songy, X., Fengx, F., Liuy, J., Liy, Z., Niey, L.: NeuroStylist: neural compatibility modeling for clothing matching. In: Proceedings of the ACM International Conference on Multimedia, June–May 2017. ACM (2017)Google Scholar
  13. 13.
    Lin, Y., Wang, T.: Dress up like a stylist? Learning from a user-generated fashion network. In: KDD 2017 Workshop on Machine Learning Meets Fashion (2017)Google Scholar
  14. 14.
    Al-Halah, Z., Stiefelhagen, R., Grauman, K.: Fashion forward: forecasting visual style in fashion. arXiv:1705.06394 (2017)
  15. 15.
    Liu, Z., Luo, P., Qiu, S., Wang, X., Tang, X.: DeepFashion: powering robust clothes recognition and retrieval with rich annotations. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)Google Scholar
  16. 16.
    Liu, Z., Yan, S., Luo, P., Wang, X., Tang, X.: Fashion landmark detection in the wild. In: European Conference on Computer Vision (ECCV) (2015)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Moscow Institute of Physics and TechnologyMoscowRussian Federation
  2. 2.Lomonosov Moscow State UniversityMoscowRussian Federation

Personalised recommendations