Humans Shouldn’t Have to Watch, Listen and Annotate Video Manually
It supports all the leading AI tech.
Real-time actions to take place without waiting for delays that occur when video is sent to the cloud.
Conextualizing and objectifying the video, still images, audio, etc. at the edge.
By enriching the media with AI-detected objects the fundamental video stream is able to be indexed and searched at a massive scale thus reducing the costs of storage and time spent manually rewinding videos, inherently eliminate the risk of humans mistakes.
Extendable AI pipeline, models, and even at infrastructure level with superb power efficiency.
Tensorflow, PyTorch & Edge Impulse.
Following the mp4 specification with
user defined box in the mp4 file
o.mp4 is an extension with a UUID
box contains context and metadata
Increased indexing and searchability.
Standardizing the extracted object
data into existing media ecosystem.
OCam is a part of the PhysicO family.
Send a Slack or instant message to an assistant when a package or person arrives.
Train for known cars and raise an alert when someone parks.
Trigger an alert and/or lights when someone not known arrives in an area.
Release new food when Fido comes in.
Trigger a workflow when someone says/yells “Help”.
Measure what percent of workers are wearing protective equipment.
How long does someone stand in front of certain products.
Detect noise volume after certain times of the day.