Abstract | ||
---|---|---|
We present a system for activity recognition from passive RFID data using a deep convolutional neural network. We directly feed the RFID data into a deep convolutional neural network for activity recognition instead of selecting features and using a cascade structure that first detects object use from RFID data followed by predicting the activity. Because our system treats activity recognition as a multi-class classification problem, it is scalable for applications with large number of activity classes. We tested our system using RFID data collected in a trauma room, including 14 hours of RFID data from 16 actual trauma resuscitations. Our system outperformed existing systems developed for activity recognition and achieved similar performance with process-phase detection as systems that require wearable sensors or manually-generated input. We also analyzed the strengths and limitations of our current deep learning architecture for activity recognition from RFID data. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1145/2994551.2994569 | SenSys |
Keywords | Field | DocType |
Activity recognition,convolutional neural network,deep learning,passive RFID,process phase detection | Activity recognition,Wearable computer,Computer science,Convolutional neural network,Speech recognition,Artificial intelligence,Deep learning,Machine learning,Scalability | Conference |
Volume | Citations | PageRank |
2016 | 21 | 0.72 |
References | Authors | |
23 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xinyu Li | 1 | 88 | 37.72 |
Yanyi Zhang | 2 | 29 | 6.40 |
Ivan Marsic | 3 | 716 | 91.96 |
Aleksandra Sarcevic | 4 | 182 | 26.75 |
Randall S. Burd | 5 | 122 | 21.53 |