The fact that objects in the world appear in different ways depending on the scale of observation has important implications if one aims at describing them. It shows that the notion of scale is of utmost importance when processing unknown measurement data by automatic methods. In their seminal works, Witkin (1983) and Koenderink (1984) proposed to approach this problem by representing image structures at different scales in a so-called scale-space representation. Traditional scale-space theory building on this work, however, does not address the problem of how to select local appropriate scales for further analysis. This article proposes a systematic methodology for dealing with this problem. A framework is proposed for generating hypotheses about interesting scale levels in image data, based on a general principle stating that local extrema over scales of different combinations of γ-normalized derivatives are likely candidates to correspond to interesting structures. Specifically, it is shown how this idea can be used as a major mechanism in algorithms for automatic scale selection, which

The factorization method described in this series of reports requires an algorithm to track the motion of features in an image stream. Given the small inter-frame displacement made possible by the factorization approach, the best tracking method turns out to be the one proposed by Lucas and Kanade

This paper investigates the use of interest operators for feature detection and the generation of high-accuracy object measurement, modelling, texturing and visualization. Interest operators are required as a preliminary step in the surface measurement process, to find features points for matching

selection criterion that is optimal by construction because it is based on how the tracker works, and a feature monitoring method that can detect occlusions, disocclusions, and features that do not correspond to points in the world. These methods are based on a new tracking algorithm that extends previous

Recently Viola et al. [5] have introduced a rapid object detection scheme based on a boosted cascade of simple feature classifiers. In this paper we introduce a novel set of rotated haar-like features. These novel features significantly enrich the simple features of [5] and can also be calculated

This paper describes a machine learning approach for visual object detection which is capable of processing images extremely rapidly and achieving high detection rates. This work is distinguished by three key contributions. The first is the introduction of a new image representation called

In this paper, we propose a novel system, named Video Booklet, which enables efficient and natural personal video browsing and searching. In the system, firstly representative thumbnails of a collection of video segments are selected through an elaborate booklet generation approach, and then reshaped by a set of pre-trained personalized shape templates (such as circle, heart, sector, stamp, etc), consequently printed out on a real booklet or album. When users plan to browse the content of their digital video library, they can firstly browse their booklets in a manner as browsing ordinary photo albums. When they want to watch a certain segment indicated by a thumbnail in the booklet, they are able to use their camera phones or similar devices to capture the corresponding thumbnail, and send the captured image to a computer via wireless network. Thereafter, the target thumbnail is accurately located by a proposed Self-Trained Active Shape Model algorithm, and then the distortion of the captured image is corrected. Finally the Video Booklet system will automatically find the most similar thumbnail to the corrected one in the video library and begin to play the corresponding segment for the users. Thereby, Video Booklet builds a seamless bridge between digital videos and analog albums.

Abstract. In this paper, we present a novel scale- and rotation-invariant interest point detector and descriptor, coined SURF (Speeded Up Ro-bust Features). It approximates or even outperforms previously proposed schemes with respect to repeatability, distinctiveness, and robustness, yet can

effects. Each feature is represented as a densely sampled parametric manifold in a low dimensional subspace of a Hilbert space. During detection, the vector of intensity values in a window about each pixel in the image is projected into the subspace. If the projection lies sufficiently close

The modern science of networks has brought significant advances to our understanding of complex systems. One of the most relevant features of graphs representing real systems is community structure, or clustering, i. e. the organization of vertices in clusters, with many edges joining vertices