Mirrors_Framework/FSRCNN-TensorFlow
1
0
Fork 0
mirror of https://github.com/igv/FSRCNN-TensorFlow.git synced 2026-02-06 23:41:59 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
50 Commits 2 Branches 5 Tags
master
Commit Graph

50 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
igv
ea173caa52 Relicense into GPLv3 2021-04-12 12:50:13 +03:00
igv
e3015e62dc Adjustments 
- decrease stride by 1 to overlap patches
 - train only once on the same image per epoch
 - replace low quality image
 - default model is 8-0-4-1
 2021-03-07 13:16:27 +02:00
Niklas Haas
3f7b9554ea tensorflow v2 compat 2020-12-16 11:19:13 +01:00
igv
aef9256d75 Move GSL to a separate function 2020-05-11 16:15:35 +03:00
igv
85c1912989 Always use multiple processes for pre-processing 2019-10-04 08:57:17 +03:00
igv
8f30c0f221 Support for scale=1 2019-10-04 08:57:17 +03:00
igv
e94e6d6cf1 CSFM Super-Resolution 2018-10-05 13:40:02 +03:00
igv
1aa11ab0e1 Switch to Gradient Sensitive Loss 1.1 2018-09-20 12:53:17 +03:00
igv
467e4e31ab Better training dataset 2018-09-08 11:36:15 +03:00
igv
45c629e5c7 Don't divide test image into patches, don't remove colors and calculate PSNR and SSIM. 2018-08-31 16:21:39 +03:00
igv
c096b67069 gen.py: support for scaling factors 3 and 4 2018-08-25 14:11:43 +03:00
igv
09d7d07496 Replace transposed convolution with sub-pixel convolution 2018-08-23 15:29:46 +03:00
igv
1b35ecb2cd Layer for predicting local sub-band residuals 2018-05-12 12:27:10 +03:00
igv
5274a73ebe gen.py: support for shrinking/expanding layers 2017-11-26 12:02:37 +02:00
igv
c64c037f01 Local residual learning 2017-10-28 17:07:25 +03:00
igv
bd0e4dc965 Support for recursive layers 2017-10-27 14:24:53 +03:00
igv
1a8bb3d9fc Global residual learning 2017-10-24 11:15:09 +03:00
igv
cc7d2722d2 Add Charbonnier loss 2018-01-20 16:41:08 +02:00
igv
d6b13cce93 Optimizations 
Mainly to speed up preprocessing and reduce memory usage
Also get rid of expand_data.py
 2017-11-04 13:56:05 +02:00
igv
be0ef42af1 Use Wand (ImageMagick) for preprocessing if available instead of Pillow 2017-11-03 20:00:13 +02:00
igv
fa94583a94 Support for (MS-)LapSRN 2017-10-21 13:22:25 +03:00
igv
e786c07a4b gen.py: generate deconvolution passes 
Usable now, except:
    - No support for shrinking/expanding layers
    - Generated shaders for scaling factors other than 2x are not compatible with mpv, they have to be modified manually
 2017-10-07 12:52:05 +03:00
igv
9e421c1d33 Half working script for generating a shader 
Doesn't support shrinking/expanding layers and doesn't generate a deconvolution passes yet
 2017-10-06 15:44:23 +03:00
igv
4a02e34490 Support for ESPCN 2017-10-05 10:30:35 +03:00
igv
e0cf791fdf Support for different radiuses for the deconvolution input tensor 
Also append radius to the checkpoint directory name.
0.5 1.0 		2017-09-10 09:37:39 +03:00
igv
286e6cfc3b Script for sorting deconvolution filter weights 2017-09-08 15:41:49 +03:00
igv
7e828beb9a Make it possible to disable shrinking/expanding layers 2017-09-03 18:42:33 +03:00
igv
5346bcf898 Adjust deconvolution filter size for different scaling factors 0.1 0.2 2017-09-01 14:14:27 +03:00
igv
f8af70e011 Append model shape to the checkpoint directory name 2017-08-21 14:06:21 +03:00
igv
15bc798701 Option to add distortions to random low-res images 2017-07-27 14:46:28 +03:00
igv
f6b024388f Randomly scale, rotate and flip the training data 2017-07-27 14:40:32 +03:00
igv
82ed39e177 Fixes 2017-07-23 15:32:28 +03:00
Niklas Haas
bbbaabe2c0 Switch to the adam optimizer instead of SGD+momentum 
Converges faster, and doesn't get stuck forever like momentum
optimization does
 2017-07-11 03:00:49 +02:00
Niklas Haas
fad61f249b Use MS-SSIM instead of MSE (PSNR) 2017-07-11 02:59:34 +02:00
Niklas Haas
db2c179879 utils.py: dump weights in a more machine parsable way 
This dumps them as literal Haskell source code, great for writing a GLSL
generator in Haskell - but it should also be valid source code in many
other languages
 2017-07-03 00:26:32 +02:00
Niklas Haas
d5bf29d6de model.py: also dump alpha variables 
In addition, I've also fixed a bug where the last alpha was hard coded
to alpha7, instead of being dynamic relative to m. This would have
introduced a conflict when m was high enough.
 2017-07-02 23:29:28 +02:00
Drake Levy
e66845dbe0 Misc changes + make code more pythonic 2017-03-14 13:51:11 -07:00
Drake Levy
3368977b38 Fixed README spacing 2017-03-14 13:20:32 -07:00
Drake Levy
322e465ae1 README spacing 2017-03-14 13:19:38 -07:00
Drake Levy
4dfb7a8e71 Modify README links 2017-03-14 13:16:25 -07:00
Drake Levy
8a1941bad3 Added super-resolved image 2017-03-01 18:01:34 -08:00
Drake Levy
007e421178 Almost ready 2017-03-01 17:59:31 -08:00
Drake Levy
906cfe03ff README usage formatting 2017-03-01 17:40:48 -08:00
Drake Levy
9dc08303cc README image formatting 2017-03-01 17:38:49 -08:00
Drake Levy
8562f557b3 Added result images, README 2017-03-01 17:38:07 -08:00
Drake Levy
9fb554e820 README and is_train -> train 2017-03-01 17:29:13 -08:00
Drake Levy
6bdc5f5a29 README usage 2017-03-01 17:27:12 -08:00
Drake Levy
bd599621f8 README update 2017-03-01 17:25:03 -08:00
Drake Levy
da3ba8f92b Updated README and pretty much everything 2017-03-01 17:23:13 -08:00
Drake Levy
9bf8cf669a Initial commit 2017-03-01 17:16:44 -08:00