Skip to content

Pytorch based implementation of ESPCN for single image super-resolution

License

Notifications You must be signed in to change notification settings

Nhat-Thanh/ESPCN-Pytorch

Repository files navigation

[Pytorch] Real-Time Super-Resolution

Implementation of ESPCN model in Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network paper with Pytorch.

Tensorflow version: https://github.com/Nhat-Thanh/ESPCN-TF

I used Adam with optimize tuned hyperparameters instead of SGD + Momentum.

I implemented 3 models in the paper, ESPCN-x2, ESPCN-x3, ESPCN-x4

Contents

Train

You run this command to begin the training:

python train.py  --steps=300000             \
                 --scale=2                  \
                 --batch_size=128           \
                 --save-best-only=0         \
                 --save-every=1000          \
                 --save-log=0               \
                 --ckpt-dir="checkpoint/x2" 
  • --save-best-only: if it's equal to 0, model weights will be saved every save-every steps.
  • --save-log: if it's equal to 1, train loss, train metric, validation loss, validation metric will be saved every save-every steps.

NOTE: if you want to re-train a new model, you should delete all files in sub-directories in checkpoint directory. Your checkpoint will be saved when above command finishs and can be used for the next times, so you can train a model on Google Colab without taking care of GPU time limit. I trained 3 models on Google Colab in 300000 steps: Open In Colab

You can get the models here

Test

I use Set5 as the test set. After Training, you can test models with scale factors x2, x3, x4, the result is calculated by compute average PSNR of all images.

python test.py --scale=2 --ckpt-path="default"
  • --ckpt-path="default" means you are using default model path, aka checkpoint/x{scale}/FSRCNN-x{scale}.h5. If you want to use your trained model, you can pass yours to --ckpt-path.

Demo

After Training, you can test models with this command, the result is the sr.png.

python demo.py --image-path="dataset/test1.png" \
               --ckpt-path="default"            \
               --scale=2
  • --ckpt-path is the same as in Test

Evaluate

I evaluated models with Set5, Set14, BSD100 and Urban100 dataset by PSNR:

Model Set5 Set14 BSD100 Urban100
ESPCN-x2 38.2830 34.4974 34.3377 31.6791
ESPCN-x3 34.6919 31.3246 31.3524 X
ESPCN-x4 32.0646 29.2934 29.7331 27.0724

Bicubic x2-x3-x4 (top), ESPCN x2-x3-x4 (bottom).

Bicubic x2-x3-x4 (top), ESPCN x2-x3-x4 (bottom).

Source: game ZingSpeed Mobile

References

About

Pytorch based implementation of ESPCN for single image super-resolution

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published