Highly-accurate binary tiny neural network for low-power human activity recognition.
Antonio De VitaDanilo PauLuigi Di BenedettoAlfredo RubinoFrédéric PétrotGian Domenico LicciardoPublished in: Microprocess. Microsystems (2021)
Keyphrases
- highly accurate
- low power
- human activity recognition
- neural network
- activity recognition
- low cost
- power consumption
- high speed
- human activities
- high quality
- capable of producing
- single chip
- high power
- low power consumption
- vlsi architecture
- high accuracy
- vlsi circuits
- logic circuits
- non binary
- machine learning
- power dissipation
- cmos technology
- digital signal processing
- image sensor
- collaborative filtering
- data streams
- mixed signal
- reinforcement learning
- gate array