2019/07/02 by Dennis Madsen, Andreas Morel-Forster, Madsen, Dennis +9 · 1 citation
Engineering · Computer Science · #3D Shape Modeling and Analysis #Generative Adversarial Networks and Image Synthesis #Robotics and Sensor-Based Localization
paper · pdf · doi:10.48550/arxiv.1907.01414
We propose to view non-rigid surface registration as a probabilistic\ninference problem. Given a target surface, we estimate the posterior\ndistribution of surface registrations. We demonstrate how the posterior\ndistribution can be used to build shape models that generalize better and show\nhow to visualize the uncertainty in the established correspondence.\nFurthermore, in a reconstruction task, we show how to estimate the posterior\ndistribution of missing data without assuming a fixed point-to-point\ncorrespondence.\n We introduce the closest-point proposal for the Metropolis-Hastings\nalgorithm. Our proposal overcomes the limitation of slow convergence compared\nto a random-walk strategy. As the algorithm decouples inference from modeling\nthe posterior using a propose-and-verify scheme, we show how to choose\ndifferent distance measures for the likelihood model.\n All presented results are fully reproducible using publicly available data\nand our open-source implementation of the registration framework.\n