Fast Robust Tensor Principal Component Analysis via Fiber CUR Decomposition *

HanQin Cai, Zehan Chao, Longxiu Huang, Deanna Needell

ICCVW 2021 pp. 189-197

doi:10.1109/ICCVW54120.2021.00026 /iccvw/2021/cai2021iccvw-fast/

Abstract

We study the problem of tensor robust principal component analysis (TRPCA), which aims to separate an underlying low-multilinear-rank tensor and a sparse outlier tensor from their sum. In this work, we propose a fast non-convex algorithm, coined Robust Tensor CUR (RTCUR), for large-scale TRPCA problems. RTCUR considers a framework of alternating projections and utilizes the recently developed tensor Fiber CUR decomposition to dramatically lower the computational complexity. The performance advantage of RTCUR is empirically verified against the state-of-the-arts on the synthetic datasets and is further demonstrated on the real-world application such as color video background subtraction.

PDF ICCVW Semantic Scholar

Cite

Text

Cai et al. "Fast Robust Tensor Principal Component Analysis via Fiber CUR Decomposition *." IEEE/CVF International Conference on Computer Vision Workshops, 2021. doi:10.1109/ICCVW54120.2021.00026

Markdown

[Cai et al. "Fast Robust Tensor Principal Component Analysis via Fiber CUR Decomposition *." IEEE/CVF International Conference on Computer Vision Workshops, 2021.](https://mlanthology.org/iccvw/2021/cai2021iccvw-fast/) doi:10.1109/ICCVW54120.2021.00026

BibTeX

@inproceedings{cai2021iccvw-fast,
  title     = {{Fast Robust Tensor Principal Component Analysis via Fiber CUR Decomposition *}},
  author    = {Cai, HanQin and Chao, Zehan and Huang, Longxiu and Needell, Deanna},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2021},
  pages     = {189-197},
  doi       = {10.1109/ICCVW54120.2021.00026},
  url       = {https://mlanthology.org/iccvw/2021/cai2021iccvw-fast/}
}