Fast Robust Tensor Principal Component Analysis via Fiber CUR Decomposition
This work addresses the challenge of efficiently separating low-rank and sparse components in large-scale tensor data, such as in video processing, though it appears incremental as it builds on existing TRPCA methods with a focus on speed.
The paper tackles the problem of tensor robust principal component analysis (TRPCA) by proposing a fast non-convex algorithm called Robust Tensor CUR (RTCUR), which uses tensor Fiber CUR decomposition to reduce computational complexity and achieves performance advantages over state-of-the-art methods on synthetic datasets and real-world applications like color video background subtraction.
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.