MarketWatch, May 4, 2021. Caspar.AI uses the whole home as a sensor – already existing elements like sensors on switches and appliances. From automated control of lights, shades and temperature to more complex tasks such as fall detection, the Caspar system works without cameras and without “wearables” such as resident pendants.
Watch-Bot: Unsupervised Learning for Reminding Humans of Forgotten Actions, ICRA, 2016.
An average adult forgets three key facts, chores or events every day in a home. To detect the action structure in the complex activities, we propose a compact yet expressive representation
that draws parallels to document modeling in the natural language. The activity is represented as a sequence of human and object “words” linked to human-skeleton and object interaction “trajectories”.