Deep generative cross-modal on-body accelerometer data synthesis from videos

Keyphrases

• Accelerometer Data 100%
• Building Model 20%
• Conditional Generative Adversarial Network (cGAN) 40%
• Conditional Variational Autoencoder 20%
• Cross-modal 100%
• Data Synthesis 100%
• Eating Recognition 20%
• Entertainment 20%
• Healthcare 20%
• Human Activity Recognition 40%
• Labeled Data 20%
• Machine Learning Models 20%
• On-body Sensors 100%
• Physical Activity Recognition 20%
• Popular 20%
• Real-time Sensor Data 20%
• Real-world Deployment 20%
• Recognition Dataset 20%
• Recognition-based 20%
• Sensor Data 20%
• Sensor-based Activity Recognition 40%
• Synthetic Data 20%
• Video Data 40%
• Wearable Sensors 40%

Engineering

• Autoencoder 20%
• Body Sensor 100%
• Data Stream 40%
• Model Building 20%
• Sensor Data 80%
• Wearable Sensor 40%