Learning Robust Representations for Computer Vision
Abstract
Unsupervised learning techniques in computer vision of ten require learning latent representations, such as low-dimensional linear and non-linear subspaces. Noise and outliers in the data can frustrate these approaches by obscuring the latent spaces. Our main goal is deeper understanding and new development of robust approaches for representation learning. We provide a new interpretation for existing robust approaches and present two specific contributions: a new robust PCA approach, which can separate foreground features from dynamic background, and a novel robust spectral clustering method, that can cluster facial images with high accuracy. Both contributions show superior performance to standard methods on real-world test sets.
Cite
Text
Zheng et al. "Learning Robust Representations for Computer Vision." IEEE/CVF International Conference on Computer Vision Workshops, 2017. doi:10.1109/ICCVW.2017.211Markdown
[Zheng et al. "Learning Robust Representations for Computer Vision." IEEE/CVF International Conference on Computer Vision Workshops, 2017.](https://mlanthology.org/iccvw/2017/zheng2017iccvw-learning/) doi:10.1109/ICCVW.2017.211BibTeX
@inproceedings{zheng2017iccvw-learning,
title = {{Learning Robust Representations for Computer Vision}},
author = {Zheng, Peng and Aravkin, Aleksandr Y. and Thiagarajan, Jayaraman J. and Ramamurthy, Karthikeyan Natesan},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2017},
pages = {1784-1791},
doi = {10.1109/ICCVW.2017.211},
url = {https://mlanthology.org/iccvw/2017/zheng2017iccvw-learning/}
}