The Video Registration, Object Tracking, and 3D Visualization modules provide analytical capabilities for automatically detecting, recognizing, and understanding important activities and events present in a given video clip. A fourth module is needed to assist humans in interpreting and analyzing the results of the clips. Video analytics provides functionality for extracting meaningful statistics and answering important questions related to the video. Using the example of a kickoff play in American football, statistics such as average hang time and kick distance can be automatically extracted, thereby eliminating manual data gathering and annotation. Video analytics provides the foundation for practical applications such as matching and retrieval of videos b ased on trajectory similarity, activity recognition, and detecti on of unusual events. Currently, t he module supports matching and retrieval of videos based on a single input trajectory using D TW matching. Continuing the kickoff play example, one could search through all kickoff plays for any number of characteristics such as characteristics, trajectories, etc. Whether in the sports, surveillance, or broadcasting domain, video analytics will help analyze, summarize, navigate, and retrieve video far more efficiently than any current methods allow.