Mohamad Zeina

I am currently training on the entire dataset, and not performing any validation. There should be a separate validation dataset, either of me at a different angle/with different lighting, or...

This should significantly improve accuracy, as some techniques rely purely on this pupil localisation with little or no calibration.

Currently, 10s of MBs have to be downloaded just to use the 6 KB script. The bulk of this extra data is from example files, which end users don't necessarily...

At the moment, this starts by deleting all existing compositing nodes. Alternatively, it could look for the "render layers" node, and insert denoising immediately after this without deleting any nodes.

Are you open to performance focussed pull requests?

1

Hey! I use this a lot on my YouTube channel Moby Motion, and would love to improve the performance for my own learning, and to help me in producing videos....