Semantics-Guided Data Hallucination for Few-Shot Visual Classification

Chia-Ching Lin, Yu-Chiang Frank Wang, Chin-Laung Lei, Kuan-Ta Chen

Abstract

Few-shot learning (FSL) addresses learning tasks in which only few samples are available for selected object categories. In this paper, we propose a deep learning framework for data hallucination, which overcomes the above limitation and alleviate possible overfitting problems. In particular, our method exploits semantic information into the hallucination process, and thus the augmented data would be able to exhibit semantics-oriented modes of variation for improved FSL performances. Very promising performances on CIFAR-100 and AwA datasets confirm the effectiveness of our proposed method for FSL.
BibTeX
@inproceedings{lin2019semantics,
  author = {Chia-Ching Lin and Yu-Chiang Frank Wang and Chin-Laung Lei and Kuan-Ta Chen},
  title = {{Semantics-Guided Data Hallucination for Few-Shot Visual Classification}},
  booktitle = {ICIP},
  year = {2019},
  doi = {10.1109/icip.2019.8803420},
}

Loading related content...

Figure