This is an attempt at describing the grammar for .itc files used by iTunes. I based this on ITC Extractor by Simon Kennedy, which does the job of extracting image data from .itc files handily.
This is just a quick & dirty attempt to highlight the basic structure of itc files. YMMV, as it's not clear whether the format of these files will change over time.
##Working with it You can apply this grammar to .itc files with Synalyze It. Occasional bugs aside, it's a good hex editor for working with & visualizing this sort of information.
I haven't found much information about this file format.
The grammar is fairly straightforward, aside from understanding the padding provided by some parts of the data. So this grammar might not be exhaustive, but it seems to be sufficient to get at the image data.
This is part of a larger project for me, but the trickiest part of working with ITC files is in dealing with ARGb data and converting it to a more useful format in native code. Since I'm only working with iTunes 11+ data and not concerned about backwards-compatibility, I haven't spent much time worrying about non-ARGb image formats.