Measuring physiological signals from the human face and body using cameras is an emerging topic that has grown rapidly in the last decade. It has been shown that image/video processing methods can be used to extract human vital signs (e.g. heart rate, heart rate variability, breathing rate, blood oxygenation, pulse transit time) from an image sequence registering the human skin or body. The use of cameras also enables the measurement of human context/semantic monitoring (e.g. emotions, facial expressions) that facilitates human understanding and heath monitoring. This will lead to various camera-based health applications that can directly improve upon contact-based clinical solutions and impact people’s quality of life (e.g. tele-medicine, care units in hospital, free-living healthcare, and fitness cardio training). Imaging methods for recovering vital signs also present new possibilities for computer vision applications that require deep understanding of human physiology, such as affective computing. However, the lack of sessions/workshops on the topic of camera-based physiological measurement in top computer vision conferences is notable and has limited the development of methods that leverage some of the latest advances in computer vision and machine learning (e.g., deep learning, adversarial networks, etc.).
The CVPM workshop aims to unite the researchers working in this field, and those who can directly/indirectly benefit from and/or contribute to it (including computer vision and machine learning researchers, doctors/clinicians, medical experts and psychologists). Although targeted at computer vision audiences, and aimed at promoting advancements in methods, a unique aspect of this workshop is that it brings a rich set of compelling applications (e.g., from video health monitoring to affective computing to face anti-spoofing and biometric security) that attracts broader audiences from fields beyond computer science.