The content can easily search labels and discover within the entire video, shot by shot, or per frame. It allows us to identify automatically key points in video content with a massive recommendation library of 20,000 labels. It also returns confidence levels for each entity identified, so it can easily control and filter video content for what’s most relevant. It provides insights on the changes that occurred in a scene or frame, and identifies appropriate locations (either regional or global) in videos to insert ads that are contextually relevant to the video content. It also automatically transcribes video content in English.
The architecture for Cloud Video Intelligence is as follows:
The core components used ...
