Parallel Structure from Motion from Local Increment to Global Averaging

Abstract

In this paper, we tackle the accurate and consistent Structure from Motion (SfM) problem, in particular camera registration, far exceeding the memory of a single computer in parallel. Different from the previous methods which drastically simplify the parameters of SfM and sacrifice the accuracy of the final reconstruction, we try to preserve the connectivities among cameras by proposing a camera clustering algorithm to divide a large SfM problem into smaller sub-problems in terms of camera clusters with overlapping. We then exploit a hybrid formulation that applies the relative poses from local incremental SfM into a global motion averaging framework and produce accurate and consistent global camera poses. Our scalable formulation in terms of camera clusters is highly applicable to the whole SfM pipeline including track generation, local SfM, 3D point triangulation and bundle adjustment. We are even able to reconstruct the camera poses of a city-scale data-set containing more than one million high-resolution images with superior accuracy and robustness evaluated on benchmark, Internet, and sequential data-sets.

Under review at the International Conference on Computer Vision (ICCV) 2017

References in corpus (3)

• Hierarchical structure-and-motion recovery from uncalibrated images
• Linear Global Translation Estimation with Feature Tracks
• Robust Camera Location Estimation by Convex Programming

Cited by in corpus (12)

• Parallel Structure from Motion for UAV Images via Weighted Connected Dominating Set
• Bundle Adjustment Revisited
• Image Based Camera Localization: an Overview
• Matchable Image Retrieval by Learning from Surface Reconstruction
• Stochastic Bundle Adjustment for Efficient and Scalable 3D Reconstruction
• Hybrid Rotation Averaging: A Fast and Robust Rotation Averaging Approach
• KFNet: Learning Temporal Camera Relocalization using Kalman Filtering
• Pushing the Envelope of Rotation Averaging for Visual SLAM
• On the Robustness of Multi-View Rotation Averaging
• Graph-Based Parallel Large Scale Structure from Motion
• VS-Net: Voting with Segmentation for Visual Localization
• VALUE: Large Scale Voting-based Automatic Labelling for Urban Environments