Multi-Label Fashion Image Classification with Minimal Human Supervision

Abstract

We tackle the problem of multi-label classification of fashion images, learning from noisy data with minimal human supervision. We present a new dataset of full body poses, each with a set of 66 binary labels corresponding to the information about the garments worn in the image obtained in an automatic manner. As the automatically-collected labels contain significant noise, we manually correct the labels for a small subset of the data, and use these correct labels for further training and evaluation. We build upon a recent approach that both cleans the noisy labels and learns to classify, and introduce simple changes that can significantly improve the performance.

Cite

Text

Inoue et al. "Multi-Label Fashion Image Classification with Minimal Human Supervision." IEEE/CVF International Conference on Computer Vision Workshops, 2017. doi:10.1109/ICCVW.2017.265

Markdown

[Inoue et al. "Multi-Label Fashion Image Classification with Minimal Human Supervision." IEEE/CVF International Conference on Computer Vision Workshops, 2017.](https://mlanthology.org/iccvw/2017/inoue2017iccvw-multilabel/) doi:10.1109/ICCVW.2017.265

BibTeX

@inproceedings{inoue2017iccvw-multilabel,
  title     = {{Multi-Label Fashion Image Classification with Minimal Human Supervision}},
  author    = {Inoue, Naoto and Simo-Serra, Edgar and Yamasaki, Toshihiko and Ishikawa, Hiroshi},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2017},
  pages     = {2261-2267},
  doi       = {10.1109/ICCVW.2017.265},
  url       = {https://mlanthology.org/iccvw/2017/inoue2017iccvw-multilabel/}
}