Search For:

Displaying 1-14 out of 14 total
Image Matching and Retrieval by Repetitive Patterns
Found in: Pattern Recognition, International Conference on
By Petr Doubek, Jiri Matas, Michal Perdoch, Ondrej Chum
Issue Date:August 2010
pp. 3195-3198
Detection of repetitive patterns in images has been studied for a long time in computer vision. This paper discusses a method for representing a lattice or line pattern by shift-invariant descriptor of the repeating element. The descriptor overcomes shift ...
 
Construction of Precise Local Affine Frames
Found in: Pattern Recognition, International Conference on
By Andrej Mikulik, Jiri Matas, Michal Perdoch, Ondrej Chum
Issue Date:August 2010
pp. 3565-3569
We propose a novel method for the refinement of Maximally Stable Extremal Region (MSER) boundaries to sub-pixel precision by taking into account the intensity function in the 2x2 neighborhood of the contour points. The proposed method improves the repeatab...
 
USAC: A Universal Framework for Random Sample Consensus
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Rahul Raguram,Ondrej Chum,Marc Pollefeys,Jiri Matas,Jan-Michael Frahm
Issue Date:August 2013
pp. 2022-2038
A computational problem that arises frequently in computer vision is that of estimating the parameters of a model from data that have been contaminated by noise and outliers. More generally, any practical system that seeks to estimate quantities from noisy...
 
Unsupervised discovery of co-occurrence in sparse high dimensional data
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Ondrej Chum, Jiri Matas
Issue Date:June 2010
pp. 3416-3423
An efficient min-Hash based algorithm for discovery of dependencies in sparse high-dimensional data is presented. The dependencies are represented by sets of features co-occurring with high probability and are called co-ocsets. Sparse high dimensional desc...
 
Lost in quantization: Improving particular object retrieval in large scale image databases
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By James Philbin, Ondrej Chum, Michael Isard, Josef Sivic, Andrew Zisserman
Issue Date:June 2008
pp. 1-8
The state of the art in visual object retrieval from large databases is achieved by systems that are inspired by text retrieval. A key component of these approaches is that local regions of images are characterized using high-dimensional descriptors which ...
 
Total Recall: Automatic Query Expansion with a Generative Feature Model for Object Retrieval
Found in: Computer Vision, IEEE International Conference on
By Ondrej Chum, James Philbin, Josef Sivic, Michael Isard, Andrew Zisserman
Issue Date:October 2007
pp. 1-8
Given a query image of an object, our objective is to retrieve all instances of that object in a large (1M+) image database. We adopt the bag-of-visual-words architecture which has proven successful in achieving high precision at low recall. Unfortunately,...
 
Optimal Randomized RANSAC
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Ondřej Chum, Jiří Matas
Issue Date:August 2008
pp. 1472-1482
A randomized model verification strategy for RANSAC is presented. The proposed method finds, like RANSAC, a solution that is optimal with user-specified probability. The solution is found in time that is (i) close to the shortest possible and (ii) superior...
 
An Exemplar Model for Learning Object Classes
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Ondrej Chum, Andrew Zisserman
Issue Date:June 2007
pp. 1-8
We introduce an exemplar model that can learn and generate a region of interest around class instances in a training set, given only a set of images containing the visual class. The model is scale and translation invariant. In the training phase, image reg...
 
Object retrieval with large vocabularies and fast spatial matching
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By James Philbin, Ondrej Chum, Michael Isard, Josef Sivic, Andrew Zisserman
Issue Date:June 2007
pp. 1-8
In this paper, we present a large-scale object retrieval system. The user supplies a query object by selecting a region of a query image, and the system returns a ranked list of images that contain the same object, retrieved from a large corpus. We demonst...
 
Epipolar Geometry from Two Correspondences
Found in: Pattern Recognition, International Conference on
By Michal Perdoch, Jiri Matas, Ondrej Chum
Issue Date:August 2006
pp. 215-219
<p>A novel algorithm for robust RANSAC-like estimation of epipolar geometry (of uncalibrated camera pair) from two correspondences of local affine frames (LAFs) is presented. Each LAF is constructed from three points independently detected on a maxim...
 
Geometric Hashing with Local Affine Frames
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Ondrej Chum, Jiri Matas
Issue Date:June 2006
pp. 879-884
We propose a novel representation of local image structure and a matching scheme that are insensitive to a wide range of appearance changes. The representation is a collection of local affine frames that are constructed on outer boundaries of maximally sta...
 
Randomized RANSAC with Sequential Probability Ratio Test
Found in: Computer Vision, IEEE International Conference on
By Jiří Matas, Ondřej Chum
Issue Date:October 2005
pp. 1727-1732
A randomized model verification strategy for RANSAC is presented. The proposed method finds, like RANSAC, a solution that is optimal with user-controllable probability n. A provably optimal model verification strategy is designed for the situation when the...
 
Epipolar Geometry Estimation via RANSAC Benefits from the Oriented Epipolar Constraint
Found in: Pattern Recognition, International Conference on
By Ondrej Chum, Tomás Werner, Jirí Matas
Issue Date:August 2004
pp. 112-115
The efficiency of epipolar geometry estimation by RANSAC is improved by exploiting the oriented epipolar constraint. Performance evaluation shows that the enhancement brings up to a two-fold speed-up. The orientation test is simple to implement, is univers...
 
Local Affine Frames for Wide-Baseline Stereo
Found in: Pattern Recognition, International Conference on
By Jiří Matas, Štěpán Obdržálek, Ondřej Chum
Issue Date:August 2002
pp. 40363
A novel procedure for establishing wide-baseline correspondence is introduced. Tentative correspondences are established by matching photometrically normalised colour measurements represented in a local affine frame. The affine frames are obtained by a num...
 
 1