Video capture of human behaviors: toward a Big Data approach

The lowering costs of cameras and data storage have led to an increasing volume of video data from a wide variety of sources. In this review, we analyze four sources of video data (i.e., traditional laboratory cameras, wearable cameras, public cameras, and private cameras), highlighting the strengths and limitations of each source regarding its utility for capturing human behaviors. While there will be technical and ethical challenges in using video camera data for human behavior research, we see promise in increased fidelity for assessing and analyzing various types of human behaviors, including behavioral occurrence, change, and development, and socio-ecological contexts. We encourage the judicious collection and secure storage of large-scale video data and the development of integrative video analytics for human behavior research.