Progressive Large-Scale Structure-from-Motion with Orthogonal MSTs

Cui, Hainan; Shen, Shuhan; Gao, Wei; Wang, Zhiheng

doi:10.1109/3dv.2018.00020

Cited by 15 publications

(10 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Only tracks generated from these selected match pairs were used in the proposed method, which yielded a robust method to avoid abundant or outlier tracks [37]. More feature correspondences of one match pair can generally produce a more robust estimate [8,9,37], which is also reasonable in the presence of repetitive structures because after geometric verification, the feature correspondences of a correctly overlapping match pair outnumber those of a match pair with a repetitive structure [9,35]. For very short baselines, which typically result in very small intersection angles, Θ is a suitable criterion to indicate their lengths.…”

Section: Construction Of Weighted View-graphmentioning

confidence: 99%

“…As discussed in Section 2.2, problematic match pairs must be eliminated because they have a negative influence on the reconstruction results of the incremental SfM method. One MST can be used as the input view-graph [35]. However, it is difficult to guarantee that its edges are correct, especially for image sets with problematic match pairs.…”

Section: Initializationmentioning

confidence: 99%

“…Here, | • | counts the number of vertices. The second MST is extracted from G , and this ensures there are no repeated edges between two MSTs, i.e., so-called orthogonal MSTs [25,35]. The above processes are repeated several times, and these extracted orthogonal MSTs compose a seed view-graph G seed with N seed iterations.…”

Section: Initializationmentioning

confidence: 99%

“…Second, the progressive scheme on the view-graph allows SfM to be carried out by iteratively prioritizing a subset of edges or match pairs, a strategy known as prioritizing match pairs. One maximum spanning tree was extracted from the view-graph to select the match pairs with the highest weights, which were used for an initial reconstruction, and expanded until no new images were added to the block among three consecutive iterations, implying that the block is completely built [35], alleviating the negative influence of some outliers in match pairs by carefully selecting the edges of the view-graph. However, a single tree is insufficiently robust if outliers are contained [9]; thus, it is dangerous to directly feed them into SfM pipelines, especially with a repetitive structure and very short baselines.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Progressive Structure from Motion by Iteratively Prioritizing and Refining Match Pairs

Xiao

Qingsong

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Structure from motion (SfM) has been treated as a mature technique to carry out the task of image orientation and 3D reconstruction. However, it is an ongoing challenge to obtain correct reconstruction results from image sets consisting of problematic match pairs. This paper investigated two types of problematic match pairs, stemming from repetitive structures and very short baselines. We built a weighted view-graph based on all potential match pairs and propose a progressive SfM method (PRMP-PSfM) that iteratively prioritizes and refines its match pairs (or edges). The method has two main steps: initialization and expansion. Initialization is developed for reliable seed reconstruction. Specifically, we prioritize a subset of match pairs by the union of multiple independent minimum spanning trees and refine them by the idea of cycle consistency inference (CCI), which aims to infer incorrect edges by analyzing the geometric consistency over cycles of the view-graph. The seed reconstruction is progressively expanded by iteratively adding new minimum spanning trees and refining the corresponding match pairs, and the expansion terminates when a certain completeness of the block is achieved. Results from evaluations on several public datasets demonstrate that PRMP-PSfM can successfully accomplish the image orientation task for datasets with repetitive structures and very short baselines and can obtain better or similar accuracy of reconstruction results compared to several state-of-the-art incremental and hierarchical SfM methods.

show abstract

Section: Construction Of Weighted View-graphmentioning

confidence: 99%

Section: Initializationmentioning

confidence: 99%

Section: Initializationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Progressive Structure from Motion by Iteratively Prioritizing and Refining Match Pairs

Xiao

Qingsong

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Brilakis et al [74] proposed a videogrammetric framework for acquiring 3D model of the outdoor scene, which uses a calibrated set of low-cost high resolution video cameras that is progressively traversed around the scene and aims to reconstruct a dense point-cloud model. Cui et al [75] hold that insufficient feature correspondences may break the completeness of the reconstructed scene, they proposed a progressive SFM approach to handle the completeness, robustness and efficiency issues in a unified framework. By progressively performing the feature tracking and pose estimation, the method can produce a large number of redundant correspondences that can improve quality of the reconstructed point clouds.…”

Section: B 3d Reconstructionmentioning

confidence: 99%

Parallel K Nearest Neighbor Matching for 3D Reconstruction

Cao

Xie

et al. 2019

IEEE Access

View full text Add to dashboard Cite

In recent years, a 3D reconstruction based on structure from motion (SFM) has attracted much attention from the communities of computer vision and graphics. It is well known that the speed and quality of SFM systems largely depend on the technique of feature tracking. If a big volume of image data is inputted for SFM, the speed of this SFM system would become very slow. And, this problem becomes severer for largescale scenes, which typically needs to capture several thousands of images to recover the point-cloud model of the scene. However, none of the existing methods fully addresses the problem of fast feature tracking. Brute force matching is capable of producing correspondences for small-scale scenes but often getting stuck in repeated features. Hashing matching can only deal with middle-scale scenes and is not capable of largescale scenes. In this paper, we propose a new feature tacking method working in a parallel manner rather than in a single thread scheme. Our method consists of steps of keypoint detection, descriptor computing, descriptor matching by parallel k-nearest neighbor (Parallel-KNN) search, and outlier rejecting. This method is able to rapidly match a big volume of keypoints and avoids to consume high computation time, then yielding a set of correct correspondences. We demonstrate and evaluate the proposed method on several challenging benchmark datasets, including those with highly repeated features, and compare to the state-ofthe-art methods. The experimental results indicate that our method outperforms the compared methods in both efficiency and effectiveness.

show abstract