Integrating computerized primitives and annotated video patterns: a proposed model for autism diagnosis and research.

The use of computerized, digital video as a means for interactive data capture has been suggested as an alternative to direct observation of behavior. The appeal of observational measures is that they are presumed to be less vulnerable to potential biases from informants, such as parents or teachers, and permit more individualized assessment that may be lost with the use of rating scales. As a potential tool for long-term, automated observation and analysis. In this technology review we propose one promising model for the integration of computerized primitives recognition and annotated video patterns as an approach to large-scale autism diagnosis and research.