Main feature: adds the coordinates conversion system to the cvserver.
This part of the code would be better placed on the flutter application from an architectural standpoint. I coded it on the server code as requested during our meeting from last week so that more people can interact with it if necessary.
This PR also refactors the CVserver to use the base64 image encoding (see #39) and still keep the human friendly file upload option.
Also, the config generation is now abstracted away from the computer vision part and the code for the server is documented.
@violoncelloCH and @nourguermazi should probably base their work for #43 on this new version of the server.
Now we only need to create a new python file (like semantic_segmentation.py) with a process_image function and swap its implementation when generating the config; as described in config.py file.
Fixes #40
Main feature: adds the coordinates conversion system to the cvserver.
This part of the code would be better placed on the flutter application from an architectural standpoint. I coded it on the server code as requested during our meeting from last week so that more people can interact with it if necessary.
This PR also refactors the CVserver to use the base64 image encoding (see #39) and still keep the human friendly file upload option. Also, the config generation is now abstracted away from the computer vision part and the code for the server is documented.
@violoncelloCH and @nourguermazi should probably base their work for #43 on this new version of the server. Now we only need to create a new python file (like
semantic_segmentation.py) with aprocess_imagefunction and swap its implementation when generating the config; as described inconfig.pyfile.