A Novel Robust Principal Component Analysis Algorithm of Nonconvex Rank Approximation

Table 3

Statistical information of SBMnet applied in the present tests.

Category name	Challenges	Frames	Dimension of stretch matrix
Highway	Cars moving	1700	320 × 240 × 1700
Bus station	Background objects moving away, objects stopping for a short while, and then moving away	617	360 × 240 × 617
Group campus	Clutter videos	300	320 × 240 × 300
Camera parameter	Videos with strong illumination changes	451	320 × 240 × 451
Advertisement board	Videos exhibiting dynamic background motion	598	504 × 336 × 598
Bus stop morning	Videos containing more than 3,500 frames	3766	320 × 240 × 3766
Toscana	Videos containing a limited number of frames (less than 20) with a very low framerate	6	800 × 600 × 6