Event Location: NSH 1507
Bio: Prof. Yael Moses, from the Efi Arazi School of Computer Science at the Interdisciplinary Center (IDC), Herzliya Israel, received a Ph.D. in computer science from the Weizmann Institute, Israel. Her early work concentrated on theoretical aspects of object recognition. Recently, she has been focusing on various aspects of multi-camera systems and distributed computer vision.

Abstract: Dynamic events such as family gatherings, concerts or sports events are often photographed by a group of people. The set of still images obtained this way is rich in dynamic content but lacks accurate temporal information. We propose a method for photo-sequencing — temporally ordering a set of still images taken asynchronously by a set of uncalibrated cameras. Our solution makes use of noisy matching of image features. Photo-sequencing is an essential tool in analyzing (or visualizing) a dynamic scene captured by still images. Conversely, we demonstrate the power of photo-sequencing for efficient and robust computation of feature matching. In particular, we propose an algorithm that restricts the image region in which matching can be found by using photo-sequencing and epipolar geometry. Experimental results show photo-sequencing and feature matching on several challenging collections of images taken using a number of mobile phones.

This talk will be based on joint works with Tali Dekel, Mor Dar and Shai Avidan.