Multimodal Unrolled Robust PCA for Background Foreground Separation

Markowitz, Spencer Abraham; Snyder, Corey; Eldar, Yonina C.; Minh, N.

doi:10.1109/tip.2022.3172851

Cited by 9 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Low-rank matrix recovery (LRMR) is extensively employed across several applications, notably in collaborative filtering for recommendation system (Zadeh Kashani and Hamidzadeh 2020), background subtraction in video processing (Markowitz et al 2022), robust principal component analysis (RPCA) for feature extraction (Wang et al 2022), matrix sensing (Ma, Li, and Chi 2021) and matrix completion (Tong, Ma, and Chi 2021). Mathematically, for a large-scale observation matrix Y = f (L * ), where operator f (•) denotes the sensing process, LRMR seeks to recover its underlying rank-r * low-rank matrix L * ∈ R n1×n2 .…”

Section: Introductionmentioning

confidence: 99%

LERE: Learning-Based Low-Rank Matrix Recovery with Rank Estimation

Xu,

Zhang,

et al. 2024

AAAI

View full text Add to dashboard Cite

A fundamental task in the realms of computer vision, Low-Rank Matrix Recovery (LRMR) focuses on the inherent low-rank structure precise recovery from incomplete data and/or corrupted measurements given that the rank is a known prior or accurately estimated. However, it remains challenging for existing rank estimation methods to accurately estimate the rank of an ill-conditioned matrix. Also, existing LRMR optimization methods are heavily dependent on the chosen parameters, and are therefore difficult to adapt to different situations. Addressing these issues, A novel LEarning-based low-rank matrix recovery with Rank Estimation (LERE) is proposed. More specifically, considering the characteristics of the Gerschgorin disk's center and radius, a new heuristic decision rule in the Gerschgorin Disk Theorem is significantly enhanced and the low-rank boundary can be exactly located, which leads to a marked improvement in the accuracy of rank estimation. According to the estimated rank, we select row and column sub-matrices from the observation matrix by uniformly random sampling. A 17-iteration feedforward-recurrent-mixed neural network is then adapted to learn the parameters in the sub-matrix recovery processing. Finally, by the correlation of the row sub-matrix and column sub-matrix, LERE successfully recovers the underlying low-rank matrix. Overall, LERE is more efficient and robust than existing LRMR methods. Experimental results demonstrate that LERE surpasses state-of-the-art (SOTA) methods. The code for this work is accessible at https://github.com/zhengqinxu/LERE.

show abstract

Section: Introductionmentioning

confidence: 99%

LERE: Learning-Based Low-Rank Matrix Recovery with Rank Estimation

Xu,

Zhang,

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Existing unrolling strategies in the context of RPCA are currently limited to algorithms based on convex relaxations. These include CORONA [12], refRPCA [13], and other similar works [14], [15], [16], [17]. However, they inherit from the previously mentioned drawbacks of such convex relaxations.…”

Section: Introductionmentioning

confidence: 99%

Deep Unrolling for Nonconvex Robust Principal Component Analysis

Tan,

Chaux,

Soubies

et al. 2023

2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing (MLSP)

View full text Add to dashboard Cite

We design algorithms for Robust Principal Component Analysis (RPCA) which consists in decomposing a matrix into the sum of a low rank matrix and a sparse matrix. We propose a deep unrolled algorithm based on an accelerated alternating projection algorithm which aims to solve RPCA in its nonconvex form. The proposed procedure combines benefits of deep neural networks and the interpretability of the original algorithm and it automatically learns hyperparameters. We demonstrate the unrolled algorithm's effectiveness on synthetic datasets and also on a face modeling problem, where it leads to both better numerical and visual performances.

show abstract