Invariant image descriptors and affine morphological scale-space
نویسنده
چکیده
In this research report we propose a novel approach to build interest points and descriptors which are invariant to a subclass of affine transformations. Scale and rotation invariant interest points are usually obtained via the linear scale-space representation, and it has been shown that affine invariance can be achieved by warping the smoothing kernel to match the local image structure. Our approach is instead based on the so-called Affine Morphological Scale-Space, a non-linear filtering which has been proved to be the natural equivalent of the classic linear scale-space when affine invariance is required. Simple local image descriptors are then derived from the extracted interest points. We demonstrate the proposed approach by robust matching experiments. Key-words: Interest points, local descriptors, scale-space representation, affine morphological scale-space. Descripteurs d’images invariants et scale-space affine morphologique Résumé : Dans ce rapport de recherche nous proposons une nouvelle approche pour construire des points d’intérêt et les descripteurs associés invariants par une sous-classe de l’ensemble des transformations affines. Les points d’intérêt invariants par changement d’échelle et rotation sont habituellement obtenus par représentation multi-échelle linéaire. De plus, il a été montré que l’invariance affine peut être obtenue en adaptant le noyau de lissage à la structure locale de l’image. Notre approche est, elle, basée sur le scale-space affine morphologique, un filtre non linéaire qui est l’équivalent naturel du scale-space linéaire quand une invariance affine est recherchée. Des descripteurs locaux simples sont ensuite extraits des images autour des points d’intérêt. Nous démontrons la validité de l’approche proposée par des expériences d’appariement robuste. Mots-clés : Points d’intérêt, descripteurs locaux, représentation multi-échelle, scale-space affine morphologique. Invariant image descriptors and affine morphological scale-space 3
منابع مشابه
Affine Invariant Interesting Descriptors
Component trees or level set graphs have proved to be efficient tools for the extraction of stable features in many image analysis and understanding applications. In this paper, we explore the relation between level sets as feature descriptors and curvature scale-space. While most classical image feature detectors are based on linear scale-space, affine-invariant descriptors can be related to t...
متن کاملAn improvement to the SIFT descriptor for image representation and matching
0167-8655/$ see front matter 2013 Elsevier B.V. A http://dx.doi.org/10.1016/j.patrec.2013.03.021 ⇑ Corresponding author. Tel./fax: +86 29 82667836 E-mail addresses: [email protected], liugz@xjtu Constructing proper descriptors for interest points in images is a critical aspect for local features related tasks in computer vision and pattern recognition. Although the SIFT descriptor has been p...
متن کاملIndexing Based on Scale Invariant Interest Points
This paper presents a new method for detecting scale invariant interest points. The method is based on two recent results on scale space: 1) Interest points can be adapted to scale and give repeatable results (geometrically stable). 2) Local extrema over scale of normalized derivatives indicate the presence of characteristic local structures. Our method first computes a multi-scale representati...
متن کاملFAI: Fast Affine Invariant Features Detection and Descriptor Extraction
Invariant feature detectors and descriptors are a common tool now for many computer vision tasks. Recently, fast and efficient variants such as BRISK were developed to offset the computational burden of traditional invariant descriptors such as SIFT. A common thread in invariant descriptors is affine invariance which improves matching performance in large baseline situations. In this work we de...
متن کاملImproving Performances of BoW-based Image Retrieval by Using Contextual Keypoint Descriptors
The paper reports an improved method of content-based image retrieval using a well-known method of bag-of words (BoW). Words built over descriptors of popular affine-invariant keypoint detectors (Harris-Affine and Hessian-Affine are exemplary choices) are used. What is novel, however, is the number of descriptors (i.e. the number of words) representing individual keypoints. Instead of SIFT (or ...
متن کامل