This paper addresses the problem of detection and delin-eation of interest points in images as part of an automatic image and video indexing for search by content purposes project. We propose a novel key point detector based on multiresolution contrast information. We compare this detector to the Plessey feature point detector as well as the detector introduced in the SUSAN project. As we are i...

The most effective therapy for the twin-to-twin transfusion syndrome (TTTS) in monochorionic twins is the fetoscopic laser occlusion of the anastomoses between the two fetal circulations. A complete closure of all anastomoses is a prerequisite for the success of the therapy, but the small field of view of the fetoscope makes it difficult for the surgeon to properly locate all the anastomoses. T...

We identify a new and interesting high-dimensional outlier detection problem in this paper, that is, detecting the subspaces in which given data points are outliers. We call the subspaces in which a data point is an outlier as its Outlying Subspaces. In this paper, we will propose the prototype of a dynamic subspace search system, called HOS-Miner (HOS stands for High-dimensional Outlying Subsp...

Article history: Received 12 February 2008 Accepted 26 November 2008 Available online xxxx

Interest point detectors are commonly employed to reduce the amount of data to be processed. The ideal interest point detector would robustly select those features which are most appropriate or salient for the application and data at hand. There is however a tradeoff between the robustness and the discriminance of the selected features. Whereas robustness in terms of repeatability is relatively...

In the paper, we propose an improved method to calculate interest points’ descriptors. The proposed method will satisfy descriptor invariant when image rotation.

In this work, we address the following matrix recovery problem: suppose we are given a set of data points containing two parts, one part consists of samples drawn from a union of multiple subspaces and the other part consists of outliers. We do not know which data points are outliers, or how many outliers there are. The rank and number of the subspaces are unknown either. Can we detect the outl...

An image is an artifact that depicts or records visual perception. Object Based Image Retrieval (OBIR) is a technique to find specific object from image database. Similar images can be retrieved from image database by comparing image features. A feature is a piece of information which is relevant for solving the computational task related to a certain application. Visual content of the entire i...