Skeletal graphs for efficient structure from motion

By Noah Snavely, S. M. Seitz, Richard Szeliski. 2008

334 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Bundle Adjustment in the Large (2010, background): “, a sparse core of images) and a set of leaf images [1].”
MatchMiner: Efficient Spanning Structure Mining in Large Image Collections (2012, methodology): “Techniques such as skeletal graphs [1] and dominating sets [3] attempt to sparsify the graph while retaining important structures.”
Building Rome in a day (2009, background): “Thus, it is preferable to find and reconstruct a minimal subset of photographs that capture the essential geometry of the scene (called a skeletal set in Snavely et al.(19)).”
Building Rome on a Cloudless Day (2010, methodology): “Aiming at scalability, Snavely et al. [ 16 ] construct skeletal sets of images whose reconstruction approximates the full reconstruction of the whole dataset.”
Towards Linear-Time Incremental Structure from Motion (2013, background): “Although a large photo collection can be reduced to a subset of iconic images [8, 5] or skeletal images [15, 2] for reconstruction, the commonly known O(n(4)) cost of the incremental SfM is still high for large scene components.”
Structure-from-Motion Revisited (2016, background): “While sparse direct methods reduce the complexity by a large factor for sparse problems, they are prohibitive for large unstructured photo collections due to typically much denser connectivity graphs [54, 4].”
Recent developments in large-scale tie-point matching (2016, methodology): “Skeletal graphs Match graphs were first explored as a tool to speed up orientation rather than matching [40].”
SfM with MRFs: Discrete-Continuous Optimization for Large-Scale Structure from Motion (2013, neutral): “One way to reduce the cost is by pruning the image set; recent work has used clustering or graph-based techniques to reduce the number of images that must be considered in SfM [4], [11], [12], [13].”
A Survey on Structure from Motion (2017, methodology): “Also, the skeletal set is required to satisfy the constraint of being uniquely realizable, which is modeled and controlled via a structure termed as the pair graph in [67].”
Global Motion Estimation from Point Matches (2012, methodology): “In the first test we used the focal lengths and radial distortion corrections produced in [6].”
Multi-View Stereo: A Tutorial (2015, methodology): “Further work on simplifying the connectivity graph of the tracks [172] and parallelization [25, 64] lead to the current state-ofthe-art SfM pipelines used in the industry, for example, Microsoft’s photosynth [16] and Google’s photo tours [15] (See Figure 1.”
Fusion of depth maps with multiple scales (2011, background): “Current structure-from-motion systems [Snavely et al. 2008] are able to recover the camera parameters of thousands of photographs under very uncontrolled conditions.”
WarpNet: Weakly Supervised Matching for Single-View Reconstruction (2016, methodology): “Given a collection of images, several heuristics have been proposed for selecting the right subset or order for multiview rigid-body reconstruction [24, 25].”