Deploy on your server:
- change gypsum.json to your configuration
- git clone the repo and put it to the processor_path on your server
- create tmp/scripts folder on your processor_path
- write your own template_path
You can also create multiple configurations, remember to change registeredServerConfig.json to your configuration file.
Run on your local machine:
- follow the following commands to create a conda environment
- conda create -n your_env_name python=3.8
- conda activate your_env_name
- pip install PyQt5
- pip install numpy
- pip install Pillow
- pip install opencv-python
- pip install opencv-contrib-python
- pip install decord
- pip install pexpect
- brew install rsync / sudo apt install rsync
- To run the program, follow the following commands
- conda activate your_env_name
- go to the root folder
- python ImageBrowser.py
Transformer module specific required packages:
- optical flow:
- PWC-NET: pytorch, cupy (pip install cupy-cuda111), flow_viz (pip install flow_viz)
- RAFT: scipy
Add your own image transformation modules:
- create your own class derived from Transform_base class
- implement getArgParser and processImageCollection methods in your own class
- register your class to registeredTransformer.json