User activity recognition and logging in distributed Intelligent Gadgets

We present userpsilas activity recognition and logging using Intelligent Gadget. Personal life logging is very useful for just diary and even memory augmentation and activity is one of key items for logging. Intelligent Gadget is a smart object for personal life logging, and has limited resources in processing and power. So we propose a simple architecture for activity recognition and logging in Intelligent Gadget which threshold-based uses activity state transition algorithm and 2-phase activity recognition. Also we show feasible services using activity recognition and logging.