DeepSense: A unified deep learning framework for time-series mobile sensing data processing

Keyphrases

• Accelerometer-gyroscope 50%
• Activity Recognition 50%
• Biometrics 50%
• Car Tracking 50%
• Challenging Tasks 50%
• Classification Applications 50%
• Classification Framework 50%
• Convolutional Neural Network 50%
• Customization 50%
• Deep Learning Framework 100%
• Displacement-based 50%
• Embedded Devices 50%
• Energy Consumption 50%
• Estimation Framework 50%
• Global Interaction 50%
• Heterogeneous Users 50%
• Human Activity Recognition 50%
• Local Interaction 100%
• Low Latency 50%
• Magnetometer 50%
• Mobile Application 50%
• Mobile Computing Application 50%
• Mobile Sensing Applications 50%
• Mobile Sensor Data 100%
• Mobile Sensors 50%
• Motion Analysis 50%
• Motion Sensor 50%
• Noise Model 50%
• Noise Pattern 50%
• On-device 50%
• Physical Noise 50%
• Physical System Models 100%
• Recurrent Neural Network 50%
• Robust Features 50%
• Rotation Rate 50%
• Sensor Input 50%
• Sensor Measurement 50%
• Sensor Noise 50%
• Sensory Modality 50%
• Series-input 50%
• Signal Classification 50%
• Signal Estimation 50%
• Signaling Dynamics 50%
• Smartphone Device 50%
• State-of-the-art Techniques 50%
• Temporal Relations 50%
• User Behavior 50%
• User Identification 50%

Computer Science

• Activity Recognition 50%
• Biometrics 50%
• Classification Framework 50%
• Convolutional Neural Network 50%
• Customisation 50%
• Data Processing 100%
• Deep Learning 100%
• Embedded Device 50%
• Energy Consumption 50%
• Human Activity Recognition 50%
• Learning Framework 100%
• Mobile Application 50%
• Mobile Sensor 50%
• Motion Analysis 50%
• Physical System 100%
• Recurrent Neural Network 50%
• Sensor Measurement 50%
• Sensory Modality 50%
• Smartphone Device 50%
• Temporal Relationship 50%
• User Behavior 50%