Local Features, All Grown Up

Vedaldi, Andrea; Soatto, Stefano

doi:10.1109/cvpr.2006.176

Cited by 29 publications

(24 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea of determining when two regions are similar up to a similarity transformation has been widely explored in the past to solve several tasks including object recognition, structure from motion, wide-baseline matching, and motion tracking [26,8,[27][28][29]. In this paper, however, we will exploit the same idea of matching similarity-invariant regions for the purpose of image denoising.…”

Section: Scale-invariant Feature Descriptorsmentioning

confidence: 99%

Nonlocal Similarity Image Filtering

Lou

Favaro

Soatto

et al. 2009

Image Analysis and Processing – ICIAP 2009

View full text Add to dashboard Cite

Abstract. We exploit the recurrence of structures at different locations, orientations and scales in an image to perform denoising. While previous methods based on "nonlocal filtering" identify corresponding patches only up to translations, we consider more general similarity transformations. Due to the additional computational burden, we break the problem down into two steps: First, we extract similarity invariant descriptors at each pixel location; second, we search for similar patches by matching descriptors. The descriptors used are inspired by scale-invariant feature transform (SIFT), whereas the similarity search is solved via the minimization of a cost function adapted from local denoising methods. Our method compares favorably with existing denoising algorithms as tested on several datasets.

show abstract

Section: Scale-invariant Feature Descriptorsmentioning

confidence: 99%

Nonlocal Similarity Image Filtering

Lou

Favaro

Soatto

et al. 2009

Image Analysis and Processing – ICIAP 2009

View full text Add to dashboard Cite

show abstract

“…We note that while the proposed match scanning algorithm cannot optimize all the parameters of the affine transform in every scanned point separately, its ability to refine the location of each point relative to the prediction allows the re-estimation of the affine transformation from all located correspondences together (we need at least three, since there are six degrees of freedom). Relative to previous work [6,15], this replaces a single non-linear six-dimensional optimization used to optimize the affine transformation estimation between two regions, with a series of N two-dimensional optimization problems that can be solved using a simple linear procedure such as NCC. Given the corrected locationsπ…”

Section: Re-estimation Of the Affine Transformmentioning

confidence: 99%

Local Region Expansion: a Method for Analyzing and Refining Image Matches

Farhan¹,

Meir²,

Hagege³

2017

Image Processing on Line

View full text Add to dashboard Cite

We present a novel method for locating large amounts of local matches between images, with highly accurate localization. Point matching is one of the most fundamental tasks in computer vision, extensively used in applications such as object detection, object tracking and structure from motion. The major challenge in point matching is to preserve large numbers of accurate matches between corresponding scene locations under different geometric and radiometric conditions, while keeping the number of false positives low. Recent publications have shown that applying an affine transformation model on local regions is a particularly suitable approach for point matching. Yet, affine invariant methods are not used extensively for two reasons: first, because these methods are computationally demanding; and second because the derived affine estimations have limited accuracy. In this work, we propose a novel method of region expansion that enhances region matches detected by any state-of-the-art method. The method is based on accurate estimation of affine transformations, which are used to predict matching locations beyond initially detected matches. We use the improved estimations of affine transformations to locally verify tentative matches in an efficient way. We systematically reject false matches, while improving the localization of correct matches that are usually rejected by state-of-the-art methods. Source CodeThe source code and documentation are available from the web page of this article 1 . The code is mainly a Matlab code that requires some Matlab toolboxes detailed in the ReadMe.txt file attached to the source code. Specific instructions on how to run the code, including some other dependencies, are also found in the ReadMe.txt file.Editorial Warning: This paper appears as a useful supplementary material and online illustration of the SIIMS paper [5]. The referees have not confirmed that the published text is an accurate or complete account of the published code.

show abstract

“…Recent works, similar in their spirit include, [28], [35] and [37]. In the cosegmentation method of [28], both images are simultaneously partitioned assuming that the common property of the foreground regions is their color probability density function (modeled by histograms) which should be also significantly different from the color distribution of the background.…”

Section: Introductionmentioning

confidence: 99%

“…In the cosegmentation method of [28], both images are simultaneously partitioned assuming that the common property of the foreground regions is their color probability density function (modeled by histograms) which should be also significantly different from the color distribution of the background. The unsupervised segmentation algorithm presented in [35], uses the best SIFT matches of Harris-Affine features to extract the common objects in image pairs. Similarly to [37], we presume that the object instances resemble in their shapes, thus having the benefit of being specific to the particular object of interest and insensitive to color and (in most cases) illumination variation.…”

Section: Introductionmentioning

confidence: 99%

Shape-Based Mutual Segmentation

2007

View full text Add to dashboard Cite

We present a novel variational approach for simultaneous segmentation of two images of the same object taken from different viewpoints. Due to noise, clutter and occlusions, neither of the images contains sufficient information for correct object-background partitioning. The evolving object contour in each image provides a dynamic prior for the segmentation of the other object view. We call this process mutual segmentation. The foundation of the proposed method is a unified level-set framework for region and edge based segmentation, associated with a shape similarity term. The suggested shape term incorporates the semantic knowledge gained in the segmentation process of the image pair, accounting for excess or deficient parts in the estimated object shape. Transformations, including planar projectivities, between the object views are accommodated by a registration process held concurrently with the segmentation.The proposed segmentation algorithm is demonstrated on a variety of image pairs. The Homography between each of the image pairs is estimated and its accuracy is evaluated.

show abstract

Local Features, All Grown Up

Cited by 29 publications

References 25 publications

Nonlocal Similarity Image Filtering

Nonlocal Similarity Image Filtering

Local Region Expansion: a Method for Analyzing and Refining Image Matches

Shape-Based Mutual Segmentation

Contact Info

Product

Resources

About