Name
Abstract | ||
|---|---|---|
This(1) paper proposes two fast randomized algorithms for solving the robust PCA problem. Alternating direction method is a popular technique to solve robust PCA problem, its dominant cost is SVD in each iteration. As randomized SVD methods have advantages in computing singular values of large-scale matrix, we combine ADM with randomized SVD methods and obtain two randomized ADM algorithms, named Randomized ADM and Extend Randomized ADM, the former of which is faster while later is more precise. We apply our algorithms to background modeling. Experimental results show that the computational time has been reduced significantly by both algorithms while the accuracy of the model is guaranteed. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1145/3207677.3278070 | PROCEEDINGS OF THE 2ND INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND APPLICATION ENGINEERING (CSAE2018) |
Keywords | DocType | Citations |
robust PCA, alternating direction method, randomized SVD, background modeling | Conference | 1 |
PageRank | References | Authors |
0.34 | 0 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sanyang Liu | 1 | 610 | 51.41 |
| Chong Zhang | 2 | 58 | 13.85 |