Deep learning for human activity recognition: A resource efficient implementation on low-power devices.
Daniele RavìCharence WongBenny LoGuang-Zhong YangPublished in: BSN (2016)
Keyphrases
- efficient implementation
- low power
- deep learning
- human activity recognition
- low power consumption
- activity recognition
- low cost
- power consumption
- high speed
- human activities
- unsupervised learning
- single chip
- machine learning
- mental models
- weakly supervised
- embedded systems
- image processing
- object recognition
- hardware implementation
- real time
- probabilistic model
- image sensor
- learning algorithm