Indexing based on scale invariant interest points

By Krystian Mikolajczyk, Cordelia Schmid. Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001, 2001

1,374 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/ddeea66c9550f99f9a6768d4b240a9fe9957487d

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

An efficient parts-based near-duplicate and sub-image retrieval system (2004, background): “Interest points [8, 13, 21] are commonly employed in a number of real-world applications such as object recognition [4] and image retrieval [17] because they can be computed efficiently, are resistant to partial occlusion, and are relatively insensitive to changes in viewpoint.”
Unsupervised object discovery via self-organisation (2012, methodology): “For the comparison we, selected the following detectors: • Harris-Laplace Mikolajczyk and Schmid (2001).”
SURF: Speeded Up Robust Features (2006, methodology): “The detector is based on the Hessian matrix [11, 1], but uses a very basic approximation, just as DoG [2] is a very basic Laplacian-based detector.”
Author Manuscript, Published in "ieee Conference on Computer Vision and Pattern Recognition on the Burstiness of Visual Elements (neutral): “It is similar to [12], where each point votes only once for an image in the database, i.e., for each database image only the best match associated with a query descriptor is kept: a descriptor cannot vote several times for the same database image.”
Visual Place Recognition: A Survey (2016, methodology): “For example, Ho and Newman [84] used Harris affine regions [85], Murillo et al. [86] and Cummins and Newman [87] Fig.”
A SIFT descriptor with global context (2005, background): “Schmid and Mohr [16] detect key points with the Harris detector and describe them using local differential graylevel invariants.”
Evaluation of Interest Point Detectors and Feature Descriptors for Visual Tracking (2011, positive): “Mikolajczyk and Schmid (2001) proposed an approach to make the Harris detector scale invariant.”
A performance evaluation of local descriptors (2005, background): “Local photometric descriptors computed for interest regions have proved to be very successful in applications such as wide baseline matching [37, 42], object recognition [10, 25], texture Corresponding author is K. Mikolajczyk,km@robots.ox.ac.uk.”
Approximate Correspondences in High Dimensions (2006, background): “For example, in computer vision, images are often represented as collections of local part descriptions extracted from regions or patches (e.g., [10, 12 ]), and many recognition algorithms rely on establishing the correspondence between the parts from two images to quantify similarity between objec”
Picking the best DAISY (2009, background): “When combined with interest points [11, 12], it facilitates point matching without initialization for image stitching and structure from motion [3, 16, 19].”
Description of interest regions with local binary patterns (2009, background): “Local features have performed very well in many computer vision applications, such as image retrieval [1], wide baseline matching [2], object recognition [3], texture recognition [4], and robot localization [5].”
On Space-Time Interest Points (2005, positive): “Here, we used the scale-adapted Harris interest point detector (Mikolajczyk and Schmid, 2001) that detects maxima of Hsp (4) in space and extrema of normalized Laplacian operator over scales (Lindeberg, 1998).”
Video Acceleration Magnification (2017, background): “The σ parameter of the Gaussian allows for selecting the observation scale of the frequency to magnify [15, 19].”
A Comparison of Affine Region Detectors (2005, background): “The estimation of aﬃne shape can be applied to any initial point given that the determinant”
Learning to Rank 3D Features (2014, background): “As mentioned above and more in [9,23,20], there are many methods to detect salient points on a scene or an object.”
Key.Net: Keypoint Detection by Handcrafted and Learned CNN Filters (2019, methodology): “9 HarrisLaplace [34] + HardNet [40] 23.”
Machine Learning for High-Speed Corner Detection (2006, methodology): “Scale-space techniques have also been combined with the Harris approach in [15] which computes Harris corners at multiple scales and retains only those which are also optima of the LoG response across scales.”
Learning Compact Binary Descriptors with Unsupervised Deep Neural Networks (2016, background): “A feature descriptor is desired to capture important and distinctive information in an image [26, 28] and also to be robust to various image transformations [26, 27].”
A Survey on Object Detection in Optical Remote Sensing Images (2016, neutral): “Many algorithms have been proposed for keypoint detection such as Harris-Laplacian detector (Mikolajczyk and Schmid, 2001) and difference of Gaussian (DoG) detector (Lowe, 2004).”