Based on proprietary Neural Network trained with real world video data from the internet and fusing spatio-temporal information in multiple frames, DeepField-SR produces superior high resolution video quality.

As DeepField-SR is designed in scalable architecture, it can flexibly respond to various high resolution upscale request. Its runtime performance on single EC2 f1.2xlarge is 11-14fps to upscale video up to 4K resolution. The API is integrated within an FFmpeg workflow, meaning that simple command enables DeepField-SR acceleration and upscaling user input video.