Tracks selection for robust, efficient and scalable large-scale structure from motion. (December 2017)

• Record Type: Journal Article
• Authors: Cui, Hainan; Shen, Shuhan; Hu, Zhanyi
• Is Part Of: Pattern recognition. Volume 72(2017:Dec.)
• Page Start: 341
• Page End: 354
• Publication Date: 2017-12
• Subjects: Tracks selection -- Bundle adjustment -- Structure from motion; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2017.08.002
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Three tracks selection criteria: Compactness, Accurateness, and Connectedness, are proposed to guide the selection process. Formulating the tracks selection task as finding a subset of tracks to cover multiple spanning trees of epipolar geometry graph (EG). Extensive experiments show that by inserting our tracks selection module into a SfM system, both the SfM efficiency and scalability are greatly improved. Our tracks selection module has a wide applicability, and it can be inserted into any global SfM systems. Our tracks selection method performs better than three other state-of-the-art tracks selection methods. Abstract: Currently global structure-from-motion (SfM) pipeline consists of four steps: estimating camera rotations first, then computing camera positions, triangulating tracks, and finally doing bundle adjustment. However, for large-scale SfM problems, the tracks are usually too noisy and redundant for the bundle adjustment. Thus in this work, we propose a novel fast tracks selection method to improve both efficiency and robustness of the bundle adjustment. Firstly, three selection criteria: Compactness, Accurateness, and Connectedness, are introduced, where the first two are to calculate a selection priority for each track and the third is to guarantee the completeness of scene structure. Then, to satisfy these criteria, a more informative subset of tracks is selected by covering multiple spanning trees of epipolar geometry graph. Since tracks selection acts only an intermediate step in the whole SfM pipeline, it can be in principle embedded into any global SfM pipelines. To validate the effectiveness of our tracks selection module, we insert it into a state-of-the-art global SfM system and compare it with three other selection methods. Extensive experiments show that by embedding our tracks selection module, the new SfM system performs similarly or better than the original one in terms of reconstruction completeness and accuracy, but is much more efficient and scalable for large-scale scene reconstructions. Finally, our tracks selection module is further embedded into two other global SfM systems to demonstrated its versatility.