Marius Leordeanu,
Monday, March 6, 2006
Abstract:
We present an efficient spectral method for finding consistent correspondences between two sets of features. We build the adjacency matrix M of a graph whose nodes represent the potential correspondences and the weights on the links represent pairwise agreements between potential correspondences. Correct assignments are likely to establish links among each other and thus form a strongly connected cluster. Incorrect correspondences establish links with the other correspondences only accidentally, so they are unlikely to belong to strongly connected clusters. We recover the correct assignments based on how strongly they belong to the main cluster of M, by using the principal eigenvector of M and imposing the mapping constraints required by the overall correspondence mapping (one-to-one or one-to-many). The experimental evaluation shows that our method is robust to outliers, accurate in terms of matching rate, while being several orders of magnitude faster than
existing methods.
Short Bio: Marius Leordeanu received a double BA in Mathematics and Computer Science from Hunter College of The City University of New York. From 2002 to 2003 hw worked in the vision lab at Hunter College in the area of 3D registration and modeling. Since 2003, he has been a PhD student at the Robotics Institute of Carnegie Mellon University. At CMU his main reasearch is focusing on object recognition.
No comments:
Post a Comment