https://qiita.com/morimoris/items/08ec250ceb09a0004768
I created Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network(ESPCN) using python and keras.
-
PC environment
CPU
: AMD Ryzen 5 3500 6-Core ProcessorMemory size
: 40GBGPU
: NVIDIA GeForce RTX 2060 SUPEROS
: Windows 10
-
Library environment
python
: 3.7.9tensorflow-gpu
: 2.4.1keras
: 2.4.3opencv-python
: 4.4.0.43
-
Create new folders which are
./train_data
and./test_data
Storage train data in
./train_data
and test data in./test_data
-
Learning
main.py --mode srcnn
- Evaluate
main.py --mode evaluate