super-resolution-service

• Perform 4x upscaling using ESRGAN
  • ONNX runtime to perform inferencing
  • ESRGAN .onnx converted from the original PyTorch model
• Compile and test locally with ./gradlew build
• Build and deploy container to CUDA enabled Docker host with:

  docker build -t super-resolution-service .
  docker run -p 8080:8080 --name super-resolution-service --gpus all super-resolution-service
  

TODO

• add more models