Loading Events

« All Events

  • This event has passed.

[VIRTUAL] Fall 2020 GRASP Seminar: Philippos Mordohai, Stevens Institute of Technology, “Improving 3D Reconstruction by Combining Data-driven Supervision and Conventional Wisdom”

October 2, 2020 @ 8:30 am - 9:30 am

ABSTRACT

Stereo matching, and geometric computer vision in general, were among the last areas of computer vision to benefit from machine learning. Recently, the pendulum has swung in the opposite direction and most new methods are almost entirely data driven. In this talk, I will present supervised learning approaches that address binocular and multi-view stereo matching by leveraging data with ground truth, as well as conventional wisdom in the form of constraints that have been proven effective over long periods of time with an emphasis on generalization.

Presenter

- Learn More

Philippos Mordohai is a Professor of Computer Science at Stevens Institute of Technology, which he joined in 2008. Prior to that, he earned his PhD from the University of Southern California and held postdoctoral appointments at the University of North Carolina and the University of Pennsylvania. His research interests span 3D reconstruction from images and video, 3D segmentation and recognition, perception for robotics, and active vision. In the past few years, his group has focused on synthesizing geometric and machine learning techniques to address problems in geometric computer vision. Dr. Mordohai is an associate editor for the Image and Vision Computing journal and Computer Vision and Image Understanding. He was selected for the 2020 Image and Vision Computing Associate Editor Award. He has served as area chair for CVPR, ICCV and ECCV and program co-chair of the International Conference on 3D Vision (3DV) in 2019, and will be one of the program chairs of ICCV 2025.

Details

Date:
October 2, 2020
Time:
8:30 am - 9:30 am
Event Categories:
,