Many different low-level feature detectors exist and it is widely agreed that the evaluation of detectors is important. In this paper we introduce two evaluation criteria for interest points: repeatability rate and information content. Repeatability rate evaluates the geometric stability under different transformations. Information content measures the distinctiveness of features. Different interest point detectors are compared using these two criteria. We determine which detector gives the best results and show that it satisfies the criteria well.

AbstractÐMany machine vision applications, such as compression, pictorial database querying, and image understanding, often need to analyze in detail only a representative subset of the image, which may be arranged into sequences of loci called regions-of-interest, ROIs. We have investigated and developed a methodology that serves to automatically identify such a subset of aROIs (algorithmically detected ROIs) using different Image Processing Algorithms, IPAs, and appropriate clustering procedures. In human perception, an internal representation directs top-down, context-dependent sequences of eye movements to fixate on similar sequences of hROIs (human identified ROIs). In this paper, we introduce our methodology and we compare aROIs with hROIs as a criterion for evaluating and selecting bottom-up, context-free algorithms. An application is finally discussed. Index TermsÐEye movements, scanpath theory, regions of interest identification and comparison. 1

Copyright © 2006 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee.

We describe a new method for automatic gait recognition based on analysing the symmetry of human motion using the Generalised Symmetry Operator. This approach is reinforced by the psychologistsÕ view that gait is a symmetrical pattern of motion and results show that gait can indeed be recognised by symmetry analysis.

Multi-class shape detection, in the sense of recognizing and localizing instances from multiple shape classes, is formulated as a two-step process in which local indexing primes global interpretation. During indexing a list of instantiations (shape identities and poses) is compiled constrained only by no missed detections at the expense of false positives. Global information, such as expected relationships among poses, is incorporated afterward to remove ambiguities. This division is motivated by computational efficiency. In addition, indexing itself is organized as a coarse-to-fine search simultaneously in class and pose. This search can be interpreted as successive approximations to likelihood ratio tests arising from a simple (“naive Bayes”) statistical model for the edge maps extracted from the original images. The key to constructing efficient “hypothesis tests” for multiple classes and poses is local OR’ing; in particular, spread edges provide imprecise but common and locally invariant features. Natural tradeoffs then emerge between discrimination and the pattern of spreading. These are analyzed mathematically within the model-based framework and the whole procedure is illustrated by experiments in reading license plates.

Abstract—We first investigate the meaning of “statistical methods ” for geometric inference based on image feature points. Tracing back the origin of feature uncertainty to image processing operations, we discuss the implications of asymptotic analysis in reference to “geometric fitting ” and “geometric model selection ” and point out that a correspondence exists between the standard statistical analysis and the geometric inference problem. Then, we derive the “geometric AIC ” and the “geometric MDL ” as counterparts of Akaike’s AIC and Rissanen’s MDL. We show by experiments that the two criteria have contrasting characteristics in detecting degeneracy. Index Terms—statistical method, feature point extraction, asymptotic evaluation, geometric AIC, geometric MDL. 1

Parallel pattern recognition requires great computational resources; it is NP-complete. From an engineering point of view it is desirable to achieve good performance with limited resources. For this purpose, we develop a serial model for visual pattern recognition based on the primate selective attention mechanism. The idea in selective attention is that not all parts of an image give us information. If we can attend only to the relevant parts, we can recognize the image more quickly and using less resources. We simulate the primitive, bottom-up attentive level of the human visual system with a saliency scheme and the more complex, top-down, temporally sequential associative level with observable Markov models. In between, there is a neural network that analyses image parts and generates posterior probabilities as observations to the Markov model. We test our model first on a handwritten numeral recognition problem and then apply it to a more complex face recognition problem. Our results indicate the promise of this approach in complicated vision applications.

Abstract—In this paper, a continuous valued measure for local color symmetry is introduced. The new algorithm is an extension of the successful gray value-based symmetry map proposed by Reisfeld et al. The use of color facilitates the detection of focus points (FPs) on objects that are difficult to detect using gray-value contrast only. The detection of FPs is aimed at guiding the attention of an object recognition system; therefore, FPs have to fulfill three major requirements: stability, distinctiveness, and usability. The proposed algorithm is evaluated for these criteria and compared with the gray value-based symmetry measure and two other methods from the literature. Stability is tested against noise, object rotation, and variations of lighting. As a measure for the distinctiveness of FPs, the principal components of FP-centered windows are compared with those of windows at randomly chosen points on a large database of natural images. Finally, usability is evaluated in the context of an object recognition task. Index Terms—Focus-of-attention, color vision, symmetry, saliency maps, object recognition. æ 1

We present a new method for detecting point matches between two images. The main issue is how to preserve the global consistency of individual matches. Existing methods propagate local smoothness by relaxation or do combinatorial search for an optimal solution. Our method imposes non-local constraints that should be approximately satisfied across the image. We define the "confidence" of such "soft constraints" to all potential matches. The confidence is progressively updated by "mean-field approximation". Finally, the "hard" epipolar constraint is imposed by RANSAC. Using real images, we demonstrate that our method is robust to camera rotations and zooming changes.

We present a system for object recognition in real images employing three dierent types of neural networks, which accomplish feature extraction and-classication. The main advantages of the method are its portability to dierent object domains without extensive parameter adjustments or changes in the feature extraction, and the low computational eort. This is achieved using a combination of vector quantization, principal component analysis and a network for nonlinear classication tasks. 1 Introduction Object recognition is one of the major problems in computer vision. It requires memorizing object specic knowledge and algorithms to compare this knowledge with an unknown image region. One way to store the required knowledge is to use explicit object models, e.g. using semantic networks. However, this requires extensive modelling by human programmers. Moreover, explicit models are mainly geometric, so many object properties which strongly determine their appearance like surface re...