Privacy-Preserving Action Recognition for Smart Hospitals using Low-Resolution Depth Images

Abstract

Computer-vision hospital systems can greatly assist healthcare workers and improve medical facility treatment, but often face patient resistance due to the perceived intrusiveness and violation of privacy associated with visual surveillance. We downsample video frames to extremely low resolutions to degrade private information from surveillance videos. We measure the amount of activity-recognition information retained in low resolution depth images, and also apply a privately-trained DCSCN super-resolution model to enhance the utility of our images. We implement our techniques with two actual healthcare-surveillance scenarios, hand-hygiene compliance and ICU activity-logging, and show that our privacy-preserving techniques preserve enough information for realistic healthcare tasks.

arXiv commentMachine Learning for Health (ML4H) Workshop at NeurIPS 2018 arXiv:1811.07216

References in corpus (6)

• Deep Residual Learning for Image Recognition
• NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis
• Towards Vision-Based Smart Hospitals: A System for Tracking and Monitoring Hand Hygiene Compliance
• Fast and Accurate Single Image Super-Resolution via Information Distillation Network
• Fast and Accurate Image Super Resolution by Deep CNN with Skip Connection and Network in Network
• Fully-Coupled Two-Stream Spatiotemporal Networks for Extremely Low Resolution Action Recognition

Cited by in corpus (2)

• Human Pose Estimation on Privacy-Preserving Low-Resolution Depth Images
• Self-supervision on Unlabelled OR Data for Multi-person 2D/3D Human Pose Estimation