English

Deep image representations using caption generators

Computer Vision and Pattern Recognition 2017-05-26 v1

Abstract

Deep learning exploits large volumes of labeled data to learn powerful models. When the target dataset is small, it is a common practice to perform transfer learning using pre-trained models to learn new task specific representations. However, pre-trained CNNs for image recognition are provided with limited information about the image during training, which is label alone. Tasks such as scene retrieval suffer from features learned from this weak supervision and require stronger supervision to better understand the contents of the image. In this paper, we exploit the features learned from caption generating models to learn novel task specific image representations. In particular, we consider the state-of-the art captioning system Show and Tell~\cite{SnT-pami-2016} and the dense region description model DenseCap~\cite{densecap-cvpr-2016}. We demonstrate that, owing to richer supervision provided during the process of training, the features learned by the captioning system perform better than those of CNNs. Further, we train a siamese network with a modified pair-wise loss to fuse the features learned by~\cite{SnT-pami-2016} and~\cite{densecap-cvpr-2016} and learn image representations suitable for retrieval. Experiments show that the proposed fusion exploits the complementary nature of the individual features and yields state-of-the art retrieval results on benchmark datasets.

Keywords

Cite

@article{arxiv.1705.09142,
  title  = {Deep image representations using caption generators},
  author = {Konda Reddy Mopuri and Vishal B. Athreya and R. Venkatesh Babu},
  journal= {arXiv preprint arXiv:1705.09142},
  year   = {2017}
}

Comments

ICME 2017

R2 v1 2026-06-22T19:58:50.843Z