Abstract

While machine learning approaches to visual recognition offer great promise, most of the existing methods rely heavily on the availability of large quantities of labeled training data. However, in the vast majority of real-world settings, manually collecting such large labeled datasets is infeasible due to the cost of labeling data or the paucity of data in a given domain. In this paper, we present a novel Adversarial Knowledge Transfer (AKT) framework for transferring knowledge from internet-scale unlabeled data to improve the performance of a classifier on a given visual recognition task. The proposed adversarial learning framework aligns the feature space of the unlabeled source data with the labeled target data such that the target classifier can be used to predict pseudo labels on the source data. An important novel aspect of our method is that the unlabeled source data can be of different classes from those of the labeled target data. Extensive experiments well demonstrate that models learned using our approach hold a lot of promise across a variety of visual recognition tasks on multiple standard datasets.

Downloads

BibTeX

@inproceedings{gupta2020adversarial,
  title={Adversarial Knowledge Transfer from Unlabeled Data},
  author={Gupta, Akash and Panda, Rameswar and Paul, Sujoy and Zhang, Jianming
  and Roy-Chowdhury, Amit K},
  booktitle={Proceedings of the 28th ACM International Conference on Multimedia},
  pages = {2175--2183},
  year={2020}
}

Acknowledgements

This work is partially supported by NSF grant 1724341, ONR grant N00014-18-1-2252 and gifts from Adobe. We thank Abhishek Aich for his assistance and feedback on the paper.