Vid.Supervisor is a machine-learning model that runs over a video to identify and tag behaviors.
Rinf.tech and its partners are working together to develop a solution rooted in the latest Machine Learning and video analytics trends. The main purpose is to vastly decrease the amount of time necessary for codifying a video, while improving the overall accuracy of the process.
For a visual map of reaching the end objective, we can take as example the next journey:
- System setup, user management, project definition
- Uploading videos into the System and defining codebook tags: actions/quantities/locations
- Streaming the videos for manual tagging
- Generating codebook excels
- System proposing values for tags
- User confirming/correcting tags
- Automated identification of patterns & behavior insight (human review is needed for 100% accuracy)
- Person Detection
- People Recognition
- Sewing Videos by: Person, Tag, Location etc.