In this paper, we investigate an approach for reconstructing storyline graphs from large-scale collections of Internet images, and optionally other side information such as friendship graphs. The storyline graphs can be an effective summary that visualizes various branching narrative structure of events or activities recurring across the input photo sets of a topic class. In order to explore further the usefulness of the storyline graphs, we leverage them to perform the image sequential prediction tasks, from which photo recommendation applications can benefit. We formulate the storyline reconstruction problem as an inference of sparse time-varying directed graphs, and develop an optimization algorithm that successfully addresses a number of key challenges of Web-scale problems, including global optimality, linear complexity, and easy parallelization. With experiments on more than 3.3 millions of images of 24 classes and user studies via Amazon Mechanical Turk, we show that the proposed algorithm improves other candidate methods for both storyline reconstruction and image prediction tasks.
This is a joint work with Gunhee Kim and Eric P. Xing.
Gunhee Kim is a postdoctoral researcher at Disney Research Pittsburgh. Prior to that, he received a Ph.D. degree at Computer Science Department of Carnegie Mellon University (CMU) in 2013, advised by Eric P. Xing. He earned a master’s degree under supervision of Martial Hebert in Robotics Institute, CMU in 2008. He also worked as a visiting student in Antonio Torralba's group at CSAIL, MIT and Fei-Fei Li's group at Stanford University. His principal research interest is solving computer vision and web mining problems that emerge from big visual data shared online, by developing scalable machine learning and optimization techniques.