Dissertation/Thesis Abstract

Learning to Predict Human Behavior from Video
by Felsen, Panna, Ph.D., University of California, Berkeley, 2019, 70; 13886183
Abstract (Summary)

In recent times, the field of computer vision has made great progress with recognizing and tracking people and their activities in videos. However, for systems designed to interact dynamically with humans, tracking and recognition are insufficient; the ability to predict behavior is requisite. In this thesis, we introduce various general frameworks for predict human behavior at three levels of granularity: events, motion, and dynamics. In Chapter 2, we present a system that is capable of predicting future events. In Chapter 3, we present a system that is capable of personalized prediction of the future motion of multi-agent, adversarial interactions. Finally, in Chapter 4, we present a framework for learning a representation of human dynamics that we can: 1) use to estimate the 3d pose and shape of people moving in videos, and 2) use to hallucinate the motion surrounding a single-frame snapshot. We conclude with several promising future directions for learning to predict human behavior from video.

Indexing (document details)
Advisor: Malik, Jitendra
Commitee: Efros, Alexei, Banks, Martin
School: University of California, Berkeley
Department: Electrical Engineering & Computer Sciences
School Location: United States -- California
Source: DAI-B 81/3(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Computer science, Artificial intelligence
Keywords: Predict human behavior, Video
Publication Number: 13886183
ISBN: 9781085792080
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest