Similarity Constrained Latent Support Vector Machine: An Application to Weakly Supervised Action Classification
Abstract
We present a novel algorithm for weakly supervised action classification in videos. We assume we are given training videos annotated only with action class labels. We learn a model that can classify unseen test videos, as well as localize a region of interest in the video that captures the discriminative essence of the action class. A novel Similarity Constrained Latent Support Vector Machine model is developed to operationalize this goal. This model specifies that videos should be classified correctly, and that the latent regions of interest chosen should be coherent over videos of an action class. The resulting learning problem is challenging, and we show how dual decomposition can be employed to render it tractable. Experimental results demonstrate the efficacy of the method.
Cite
Text
Shapovalova et al. "Similarity Constrained Latent Support Vector Machine: An Application to Weakly Supervised Action Classification." European Conference on Computer Vision, 2012. doi:10.1007/978-3-642-33786-4_5Markdown
[Shapovalova et al. "Similarity Constrained Latent Support Vector Machine: An Application to Weakly Supervised Action Classification." European Conference on Computer Vision, 2012.](https://mlanthology.org/eccv/2012/shapovalova2012eccv-similarity/) doi:10.1007/978-3-642-33786-4_5BibTeX
@inproceedings{shapovalova2012eccv-similarity,
title = {{Similarity Constrained Latent Support Vector Machine: An Application to Weakly Supervised Action Classification}},
author = {Shapovalova, Nataliya and Vahdat, Arash and Cannons, Kevin J. and Lan, Tian and Mori, Greg},
booktitle = {European Conference on Computer Vision},
year = {2012},
pages = {55-68},
doi = {10.1007/978-3-642-33786-4_5},
url = {https://mlanthology.org/eccv/2012/shapovalova2012eccv-similarity/}
}