At present the world is moving towards the next generation 5G networks. With the advent of 5G communication, internet traffic has reached its apex. As per 2019 Global Internet Phenomena Report , video data accounts for over 60 percent of downstream traffic on the internet and it is increasing day by day. Video has become one of the most commonly used multimedia in our daily life and as High Definition (HD) and Ultra-High Definition (UHD) display devices are gaining more attention. This facilitates the need for higher data-rates and greater spectral efficiency.
A problem with the video decompression is that we need to take care of temporal features between frames to achieve good decompression rate. We designed a new approach to handle this problems where we propose a new compressed learning network which could learn both the temporal and spatial features quickly and reconstruct and increase their resolution simultaneously.