One of the main limitations to the adoption of deep learning for image compression is the
need to train multiple models to compress at multiple rates. In the case of onboard remote …