Incremental gradient on the Grassmannian for online foreground and background separation in subsampled video

By Jun He, Laura Balzano, Arthur Szlam. 2012 IEEE Conference on Computer Vision and Pattern Recognition, 2012

391 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Streaming PCA and Subspace Tracking: The Missing Data Case (2018, neutral): “[83] J.”
Unifying Nuclear Norm and Bilinear Factorization Approaches for Low-Rank Matrix Decomposition (2013, methodology): “Matrix RPCA-IALM [24] GRASTA [19] PRMF [36] Ours N r Error Time Error Time Error Time Error Time 100 3 1.4872 0.3389 226.46 1.7656 3338.7 0.4704 0.5286 0.1734 200 5 1.5599 2.3575 241.99 2.7282 2687.5 1.0382 0.7182 0.5739 500 10 3.2595 10.501 263.55 9.5399 1692.4 6.2480 0.1273 3.2373 1000 15 0.3829 4”
Decomposition into Low-rank plus Additive Matrices for Background/Foreground Separation: A Review for a Comparative Evaluation with a Large-Scale Dataset (2015, methodology): “[143][144] subsampled At on the index set Ωt ⊂ {1, .”
Robust Subspace Learning: Robust PCA, Robust Subspace Tracking, and Robust Subspace Recovery (2017, methodology): “Grassmannian Robust Adaptive Subspace Tracking Algorithm (GRASTA) [50].”
Incremental learning algorithms and applications (2016, background): “Typical problems in this context range from object recognition [36, 9, 98], image segmentation [36, 71], and image representation [30, 165], up to video surveillance, person identification, and visual tracking [154, 28, 167, 104, 37, 134, 174, 101].”