Highlight detection and automatic video editing for online game streamers
Highlight detection and automatic video editing
Modulai has built a object-detection based system for detecting highlights and memorable events in online gaming streams for automatic editing of engaging highlight reels.
Background
A content creator that creates a highlight reel of their best moments in a multi hour stream spends hours on finding the highlights and editing down a video to a 5 minute version. For a professional top tier streamer with a team behind, although expensive and tedious, it’s managable. An automatic solution can unlock this capability to millions of streamers globaly and have Facebook, Youtube and TikTok-ready conent the instance the stream is finished.
Solution
A sophisticated pipeline for synthetic data generation and model training was developed in Python, evaluation, detection and editing solution in Python and C++ and used to onboard some of the most played game titles in the world. The system track events, evaluates their relevance, builds confidence and decides intervals to edit out and join together to a final video.
The result
The solution was used to onboard 10 of the most streamed games in the world from various genres and is capable of accurately detect and edit videos with respect to the most important and noteworthy moments in each game.
Automatically edited highlight video of Valorant, made out of an hour of material.
