0

In my research, there is a current task of small (64x64 pixels) image compression. The images are photos, represented as RGB. While it is not a core interest, I wanted to compare some compression methods with ML compression methods. I am NOT a machine learning specialist (I never trained any network) so I need something off the shelf, which I can just download and compress/decompress images. I am a decent user of Python so this part is not a problem.

Can someone advise on (in some sense) a respected and already trained NN framework that I can use for this purpose? The "respected" part is important because if I want to include these results in my scientific paper, it cannot be just a random guy on GitHub :)

Yauhen Yakimenka
  • 101
  • 1
  • 1
    Why don't you have a look at [paperswithcode](https://paperswithcode.com/task/image-compression/latest) they have a list of image compression benchmarks and papers that have open-source implementations. I didn't go through any of the papers or implementations, but it looks to me that the majority are ML-based methods. – Lars Mar 03 '22 at 14:38
  • Thanks for that! I was not really aware of this, will check it out – Yauhen Yakimenka Mar 03 '22 at 16:45
  • Yauhen Yakimenka, did you find what you were looking for? – Lars Mar 05 '22 at 07:29
  • To be honest, it takes longer than I expected to figure out which of those fit my needs... – Yauhen Yakimenka Mar 05 '22 at 11:32

0 Answers0