Understanding Singular Value Decomposition for Image Compression

[Th3HoopMan]

I have no idea what it is, and I really don't know how to go about using SVD on a matrix. All I know is that it could be useful for image compression (okay...?). Are there any resources or videos I can look at that will simplify the problem solving process for this? I've found a few that pertain to the theory but it's usually pretty confusing.

 

[mfb]

As usual, the wikipedia article has an introduction and many links to resources that go into more detail. The basic idea is to decompose the matrix into "important" and "unimportant" components in a different basis, and then drop the unimportant parts.

Which problem solving process do you mean?

 

[Th3HoopMan]

Thanks I'll take a look. As for the problems we usually get instructions telling us to compute a SVD of [inset random matrix here].

 

[mfb]

Well, there are code packages that can do that.

 

[Th3HoopMan]

Yeah i found that out but I have to do it by hand for a class I'm in