Robust Global Translations with 1DSfM

By Kyle Wilson, Noah Snavely. 2014

471 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

ASLFeat: Learning Local Features of Accurate Shape and Localization (2020, background): “(T&T) [13] for wide-baseline reconstruction, and the Community Photo Collection (CPC) [38] for wild reconstruction from web images.”
Structure-from-Motion Revisited (2016, background): “81 NYC Library [61] 2,550 339 400 411 453 66K (4.”
MAGSAC++, a Fast, Reliable and Accurate Robust Estimator (2019, methodology): “MAGSAC++ is the most accurate robust estimator on CPC, Tanks and Temples and TUM datasets since its curve is always higher than that of the other methods.”
Privacy Preserving Image Queries for Camera Localization (2019, methodology): “Dataset names [37, 74] and statistics on the left with corresponding localization errors for single-image scenario as box plots to the right.”
D2-Net: A Trainable CNN for Joint Description and Detection of Local Features (2019, methodology): “For evaluation, we use three medium-scale internet-collected datasets with a signiﬁcant number of different cameras and conditions (Madrid Metropolis, Gendarmen-markt and Tower of London [64]) from a recent local feature evaluation benchmark [52].”
A Survey on Structure from Motion (2017, methodology): “…which is composed of the detection of outliers among the pairwise direction measurements, via a procedure named 1DSfM , followed by a (non-convex) location estimation method, is studied in [83] (cf. Figure 6, courtesy of [83], for a simpliﬁed illustration of the outlier detection procedure 1DSfM).”
Comparative Evaluation of Hand-Crafted and Learned Local Features (2017, methodology): “We test the descriptors on the large-scale Internet datasets by Wilson and Snavely [53].”
LIFT: Learned Invariant Feature Transform (2016, methodology): “We used the collections from Piccadilly Circus in London and the Roman Forum in Rome from [26] to reconstruct the 3D using VisualSFM [27], which relies of SIFT features.”
OpenMVG: Open Multiple View Geometry (2016, methodology): “- Translation averaging with L2 norm [17], L1, and L∞ [5].”