Robust Nonrigid Registration by Convex Optimization

Qifeng Chen Vladlen Koltun
IEEE International Conference on Computer Vision (ICCV), 2015


We present an approach to nonrigid registration of 3D surfaces. We cast isometric embedding as MRF optimization and apply efficient global optimization algorithms based on linear programming relaxations. The Markov random field perspective suggests a natural connection with robust statistics and motivates robust forms of the intrinsic distortion functional. Our approach outperforms a large body of prior work by a significant margin, increasing registration precision on real data by a factor of 3.


Slide (PDF, PPT)
Complete code package (updated on 27 April 2017, 8.5 Gb)
Minimal code package (updated on 27 April 2017, 366 Mb)
Script to plot Figure 3