Learning from Video and Text via Large-Scale Discriminative Clustering
Abstract
Discriminative clustering has been successfully applied to a number of weakly supervised learning tasks. Such applications include person and action recognition, text-to-video alignment, object co-segmentation and colocalization in videos and images. One drawback of discriminative clustering, however, is its limited scalability. We address this issue and propose an online optimization algorithm based on the Block-Coordinate Frank-Wolfe algorithm. We apply the proposed method to the problem of weakly supervised learning of actions and actors from movies together with corresponding movie scripts. The scaling up of the learning problem to 66 feature-length movies enables us to significantly improve weakly supervised action recognition.
Cite
Text
Miech et al. "Learning from Video and Text via Large-Scale Discriminative Clustering." International Conference on Computer Vision, 2017. doi:10.1109/ICCV.2017.562Markdown
[Miech et al. "Learning from Video and Text via Large-Scale Discriminative Clustering." International Conference on Computer Vision, 2017.](https://mlanthology.org/iccv/2017/miech2017iccv-learning/) doi:10.1109/ICCV.2017.562BibTeX
@inproceedings{miech2017iccv-learning,
title = {{Learning from Video and Text via Large-Scale Discriminative Clustering}},
author = {Miech, Antoine and Alayrac, Jean-Baptiste and Bojanowski, Piotr and Laptev, Ivan and Sivic, Josef},
booktitle = {International Conference on Computer Vision},
year = {2017},
doi = {10.1109/ICCV.2017.562},
url = {https://mlanthology.org/iccv/2017/miech2017iccv-learning/}
}