You are here

Cost-Sensitive Top-Down/Bottom-Up Inference for Multiscale Activity Recognition

TitleCost-Sensitive Top-Down/Bottom-Up Inference for Multiscale Activity Recognition
Publication TypeConference Paper
Year of Publication2012
AuthorsAmer, M. R., D. Xie, M. Zhao, S. Todorovic, and S-C. Zhu
Conference Name12th European Conference on Computer Vision
Pagination187 - 200
Date Published10/2012
PublisherSpringer Berlin Heidelberg
Conference LocationFlorence, Italy
ISBN Number978-3-642-33765-9

This paper addresses a new problem, that of multiscale activity recognition. Our goal is to detect and localize a wide range of activities, including individual actions and group activities, which may simultaneously co-occur in high-resolution video. The video resolution allows for digital zoom-in (or zoom-out) for examining fine details (or coarser scales), as needed for recognition. The key challenge is how to avoid running a multitude of detectors at all spatiotemporal scales, and yet arrive at a holistically consistent video interpretation. To this end, we use a three-layered AND-OR graph to jointly model group activities, individual actions, and participating objects. The AND-OR graph allows a principled formulation of efficient, cost-sensitive inference via an explore-exploit strategy. Our inference optimally schedules the following computational processes: 1) direct application of activity detectors – called α process; 2) bottom-up inference based on detecting activity parts – called β process; and 3) top-down inference based on detecting activity context – called γ process. The scheduling iteratively maximizes the log-posteriors of the resulting parse graphs. For evaluation, we have compiled and benchmarked a new dataset of high-resolution videos of group and individual activities co-occurring in a courtyard of the UCLA campus.