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

Keyphrases

• On-body Sensors 100%
• Sensor-based Activity Recognition 40%
• Human Activity Recognition 40%
• Conditional Generative Adversarial Network (cGAN) 40%
• Recognition Dataset 20%
• Physical Activity Recognition 20%
• Real-time Sensor Data 20%
• Conditional Variational Autoencoder 20%

Engineering

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