Direct Linear Structure-from-Motion

  • Ping Tan | Simon Fraser University

Many structure-from-motion (SfM) systems take an incremental approach to reconstruct input images one by one. Global SfM methods that estimate all cameras simultaneously are highly desirable in terms of computation efficiency and result accuracy. We present a direct linear method for global SfM, which solve all cameras in a linear equation. We derive these linear equations from camera triplets, and generalize them to feature tracks to deal with weakly associated data. Experiments on various data demonstrate the superior efficiency and accuracy of our method.

Speaker Details

Ping Tan obtained his PhD in Computer Science from the Hong Kong University of Science and Technology in 2007, and his MS and BS degrees from the Shanghai Jiao Tong University in 2000 and 2003 respectively. Dr. Tan is an assistant professor in the School of Computing Science in the Simon Fraser University, before that he held positions at the National University of Singapore from 2007 to 2014. Dr. Tan’s research interests include computer vision and computer graphics. He is on the editorial board of the International Journal of Computer Vision (IJCV) and Machine Vision and Applications (MVA). He has served in the program committees of SIGGRAPH and SIGGRAPH Asia. Dr. Tan received the inaugural TR35@Singapore award in 2012, the Image and Vision Computing Outstanding Young Researcher Honorable Mention Award in 2012.

    • Portrait of Jeff Running

      Jeff Running

Series: Microsoft Research Talks