Hello!
LoLWatcher (Colab version only) has been updated! I have added this as a new version so that it does not interfere with the original deadline submitted version.
This new version:
- Converts Matplotlib image display to OpenCV image display in Colab
- Allows for bounding boxes of text to be shown
- Allows Tesseract detected text to be shown (This is what the notebook is "seeing"!)
- These changes allow for much better visual clarity and a peak behind the scenes of how the prediction value is being obtained
- However, these changes result in a much slower (Up to 4x) run per cycle. It is advised to not use this new feature on every screenshot, and instead should only be used to get an idea of where errors are occurring.
You can find the new version with this new notebook using this link
Log in or sign up for Devpost to join the conversation.