Abstract | ||
---|---|---|
We present a simple and effective image super-resolution algorithm that imposes an image formation constraint on the deep neural networks via pixel substitution. The proposed algorithm first uses a deep neural network to estimate intermediate high-resolution images, blurs the intermediate images using known blur kernels, and then substitutes values of the pixels at the un-decimated positions with those of the corresponding pixels from the low-resolution images. The output of the pixel substitution process strictly satisfies the image formation model and is further refined by the same deep neural network in a cascaded manner. The proposed framework is trained in an end-to-end fashion and can work with existing feed-forward deep neural networks for super-resolution and converges fast in practice. Extensive experimental results show that the proposed algorithm performs favorably against state-of-the-art methods. |
Year | Venue | DocType |
---|---|---|
2020 | AAAI | Conference |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jin-shan Pan | 1 | 567 | 30.84 |
Deqing Sun | 2 | 1061 | 44.84 |
Yang Liu | 3 | 0 | 0.68 |
Jimmy S. J. Ren | 4 | 324 | 23.85 |
Ming-Ming Cheng | 5 | 1914 | 82.32 |
Jian Yang | 6 | 6102 | 339.77 |
Jinhui Tang | 7 | 5180 | 212.18 |