Weakly labeled fine-grained classification with hierarchy relationship of fine and coarse labels

Qihan Jiao, Zhi Liu, Linwei Ye, Yang Wang

Abstract

Highlights•We propose a fine-grained image classification model under a weakly supervised setting.•Our model performs adaptive classification according to the available labels.•Convolutional block attention module is introduced to highlight meaningful features.•Multi-scale and multi-level features are combined for comprehensive feature representation.•Results on three datasets demonstrate the better performance of our model.
BibTeX
@article{jiao2019weakly,
  author = {Qihan Jiao and Zhi Liu and Linwei Ye and Yang Wang},
  title = {{Weakly labeled fine-grained classification with hierarchy relationship of fine and coarse labels}},
  journal = {J. Vis. Commun. Image Represent.},
  year = {2019},
  url = {https://openreview.net/forum?id=3MSQDqSqlh},
}

Loading related content...

Figure