Few Shot Learning

For every one hundred frames in the video, we only transmit one, and use adversarial learning to fill in the gaps. This allows us to save data transmission, and makes up for latency.

Super resolution

Since we only transmit one frame out of one hundred, we're able to apply super resolution techniques to make sure the video hits perfect quality.