Image Super-Resolution via Iterative Refinement · HackerPapers discussions

Image Super-Resolution via Iterative Refinement — mentioned by dagw

Wed, 01 Sep 2021 12:42:17 GMT

The original paper[1] itself accidentally demonstrates how dangerous this will be. Look at the picture of the leopard and study the patter of the spots around the face. The pattern on the upscaled image is clearly different from the original. The algorithm has just generated 'realistic' looking spots where it thinks there should be spots, and they have no relation to reality. [1] https://arxiv.org/pdf/2104.07636.pdf

Paper and discussion history. A mention is not an endorsement.

Image Super-Resolution via Iterative Refinement — mentioned by Ballas

Wed, 01 Sep 2021 06:20:38 GMT

That is indeed a shortcoming of this article in my opinion as well. If you want a comparison to the original high res photos, there are some examples in the original paper for SR3: https://arxiv.org/pdf/2104.07636.pdf Have not had a look at the CDM paper.

Paper and discussion history. A mention is not an endorsement.

Image Super-Resolution via Iterative Refinement — mentioned by AndrewKemendo

Wed, 01 Sep 2021 04:27:48 GMT

They do in the paper. I'm not sure why the article doesn't. https://arxiv.org/pdf/2104.07636.pdf

Paper and discussion history. A mention is not an endorsement.

Image Super-Resolution via Iterative Refinement — mentioned by bscphil

Fri, 16 Jul 2021 23:57:18 GMT

I see only two examples of reference images on that page, but they have quite a few more examples in the paper itself: https://arxiv.org/pdf/2104.07636.pdf The results compared with reference images pretty clearly shows that the technique isn't "magical" in any sense of the term - meaning that the ML algorithm doesn't rediscover real details that a human could not infer were there from the original. While the results are certainly incredibly good looking and an advance in the state of the art, if you were hoping to use this as the basis of an image compression codec you will probably be…

Paper and discussion history. A mention is not an endorsement.

Image Super-Resolution via Iterative Refinement — HN submission

Tue, 13 Jul 2021 15:36:16 GMT

Image Super-Resolution via Iterative Refinement

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.