Authors: Adam Sadilek and Henry Kautz
Abstract: Recent research has shown that surprisingly rich models of human activity can be learned from GPS (location) data. However, most research to date has concentrated on modeling single individuals or statistical properties of groups of individuals. In contrast, given noisy real-world GPS data, we consider the problem of modeling and recognizing multi-agent activities and the intentions of the people involved. Our test domain is the game of capture the flag (CTF), an outdoor game that involves many distinct cooperative and competitive joint activities. This paper takes on the task of understanding CTF from GPS data as an exemplar of the general problem of inferring human interactions and intentions from sensor data. We demonstrate that given raw data, we can automatically and reliably detect and recognize both successful and failed attempts at activities within this complex multi-agent domain. Additionally, we show that success, failure, goal, and intent of an activity are intimately tied together and having a model for successful events allows us to naturally learn models of the other three important aspects of life. We compare our approach with two alternatives and show that the unified model, which takes into account not only relationships among individual players, but also relationships among activities over the entire length of a game, although more computationally costly, is significantly more accurate.