The Defense Advanced Research Projects Agency's (DARPA) Mind's Eye Project seeks to generate research in computer vision and artificial intelligence. The goal of the project is to create computer systems capable of perceiving humans in action and describe the actions in conversational English. This book documents efforts to create ground truth labeling for video files that can be used as training data for computer systems, as well as efforts to organize these data into a well-structured database. Ground-truth labeling efforts include data gathered through panel studies as well as crowdsourcing efforts through the Amazon Mechanical Turk service.