I worked in a small research company that had a method (segmentation + CNNs, etc) a few years back. We had some exciting stuff also on masking effects, but as s

I worked in a small research company that had a method (segmentation + CNNs, etc) a few years back. We had some exciting stuff also on masking effects, but as soon as we got into SaMD and the main revenue stream (grants) dried up, engineering closed down.