Multi-image matching using multi-scale oriented patches

By Matthew A. Brown, Richard Szeliski, Simon A. J. Winder. Computer Vision and Pattern Recognition, 2005

472 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/128477ed907857a8ae96cfd25ea6b2bef74cf827

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

Unsupervised Discovery of Mid-Level Discriminative Patches (2012, background): “From the low-level, bottom-up point of view, an image patch simply represents the appearance at that point, either directly (with raw pixels [8]), or transformed into a diﬀerent representation (ﬁlterbank response vector [9], blurred [10,11] or spatially-binned [12,13] feature, etc).”
Infrared Small Target Detection Based on Partial Sum of the Tensor Nuclear Norm (2019, methodology): “In [63], a “corner strength” function was computed to In [63], a “corner strength” function was computed to ﬁnd the interest points: 1 2 where ( , ) x y represents the pixel location, ( ) ⋅ ST denotes the structure tensor, ( , ) ST x y is a matrix, ( ) ⋅ det and ( ) ⋅ tr are the determinant and…”
Description of Interest Regions with Center-Symmetric Local Binary Patterns (2006, unknown): “For some interesting recent work on interest region description done after this study, see [8,9,10,11].”
Unsupervised 3D object recognition and reconstruction in unordered datasets (2005, background): “This spatial accumulation is important for shift invariance, since the interest point locations are typically accurate in the 1-3 pixel range [6].”
Learning Local Image Descriptors (2007, background): “Various researchers have made use of the ratio test when determining if two descriptors are a match or non-match [3].”
ORB: An efficient alternative to SIFT or SURF (2011, methodology): “The closest system to ORB is [3], which proposes a multi-scale Harris keypoint and oriented patch descriptor.”
Dense interest points (2010, background): “Similarly, in their image stitching work, Brown et al. [2] introduce the concept of adaptive non-maximal suppression .”
Applying Deep Learning in Augmented Reality Tracking (2016, background): “Non-maximal suppression [21] is usually employed to get rid of the extraneous points but it still can cause localization errors.”