Efficient Initial Pose-graph Generation for Global SfM
This work provides a significant speedup for global Structure-from-Motion algorithms, which is beneficial for researchers and practitioners working with large image datasets in computer vision.
This paper addresses the problem of slow initial pose-graph generation in global Structure-from-Motion by proposing methods that avoid expensive FLANN and RANSAC steps. Their approach, which leverages consecutive image matching and known epipolar geometry, resulted in a 17x speedup for feature matching and a 5x speedup for pose estimation on the 1DSfM dataset.
We propose ways to speed up the initial pose-graph generation for global Structure-from-Motion algorithms. To avoid forming tentative point correspondences by FLANN and geometric verification by RANSAC, which are the most time-consuming steps of the pose-graph creation, we propose two new methods - built on the fact that image pairs usually are matched consecutively. Thus, candidate relative poses can be recovered from paths in the partly-built pose-graph. We propose a heuristic for the A* traversal, considering global similarity of images and the quality of the pose-graph edges. Given a relative pose from a path, descriptor-based feature matching is made "light-weight" by exploiting the known epipolar geometry. To speed up PROSAC-based sampling when RANSAC is applied, we propose a third method to order the correspondences by their inlier probabilities from previous estimations. The algorithms are tested on 402130 image pairs from the 1DSfM dataset and they speed up the feature matching 17 times and pose estimation 5 times.