Deep Encode: Scene based Video Encoding for live and VoD
Video streaming content differs in terms of complexity and requires title-specific encoding settings to guarantee a certain visual quality. FAMIUM Deep Encode leverages different machine learning techniques to identify the optimal encoding ladder for all types of video content. Our solution supports title- and scene-based encoding for VoD and live streams. By constant validation and retraining of our models the accuracy and efficiency of the underlying algorithms is improved with each input and the entire system learns on its own. Furthermore, it uses information from SAND metric reporting to enable context-aware encoding and meet the desired QoE expectations.