Is there a way to get wget to recursively download only embedded images from a website? What I want to do is point wget at a webpage, have it follow the links from that page and download the images embedded in the second-level pages. I don't need or want to save the html or anything else.

I'd really appreciate any advice.