I know that before running Principal component analys, the input data needs to be centered around its mean (subtract the mean from each keypoint) before running the algorithm.

Do I need to center my data before running robust PCA ?

For instance, say I have a video from a static security camera (like in the original work). Do I need to subtract from each pixel (in each frame), its mean value over the duration of the video?

p.s. any other pre-processing required for RPCA?


If you would like to standardize your variable to mean zero and standard deviation one, I would suggest:

$s_{i} = \frac{x_{i}-c_{1}}{\sqrt{c_{2}-c_{1}^{2}}}$

where $x_{i}$ is your raw (not standardized) variable, $c_{1} = \sum_{i=1}^{N}x_{i}$ and $c_{2} = \sum_{i=1}^{N}x_{i}^{2}$.

That way, $s_{i}$ would have mean zero and standard deviation one. You could obviously leave away the standardization (standard deviation one) if you are not interested in that.

Is that what you are looking for?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.