PlaneMatch: Patch Coplanarity Prediction for Robust RGB-D Reconstruction
Yifei Shi1     Kai Xu1     Matthias Nießner2     Szymon Rusinkiewicz3     Thomas Funkhouser3    
    1National University of Defense Technology (NUDT)     2Technical University of Munich     3Princeton University
Proceedings of the European Conference on Computer Vision (ECCV 2018)

We introduce a novel RGB-D patch descriptor designed for detecting coplanar surfaces in SLAM reconstruction. The core of our method is a deep convolutional neural net that takes in RGB, depth, and normal information of a planar patch in an image and outputs a descriptor that can be used to find coplanar patches from other images.We train the network on 10 million triplets of coplanar and non-coplanar patches, and evaluate on a new coplanarity benchmark created from commodity RGB-D scans. Experiments show that our learned descriptor outperforms alternatives extended for this new task by a significant margin. In addition, we demonstrate the benefits of coplanarity matching in a robust RGBD reconstruction formulation.We find that coplanarity constraints detected with our method are sufficient to get reconstruction results comparable to state-of-the-art frameworks on most scenes, but outperform other methods on standard benchmarks when combined with a simple keypoint method.