I’m currently working on a platform for processing videoclips. Our job is to extract as much information as possible (e.g. detect tags, detect faces, get dominant colors, etc.) out of each clip. Most of the heavy lifting is delegated to 3rd-party services, such as AWS Rekognition and Google Video Intelligence, which can provide the most “reliable” and cheap Machine Learning platform for this kind of tasks.
Without going too much into details, target videoclips …Read More