Researches of image-based 3D reconstruction have recently produced a number of good results, but they assume that the accurate foreground to be reconstructed is already extracted from each input image. This paper proposes a novel approach to extract more accurate foregrounds by iteratively performing foreground extraction and 3D reconstruction in a manner similar to an EM algorithm on regions segmented in an initial stage, called segments. After definitively extracting the foregrounds in multi-views based on simply selecting segments corresponding to the real foreground in only one image, further improved foregrounds are extracted by back-projecting 3D objects reconstructed based on the foreground extracted in the previous step into segments of each image in multi-views. These two steps are iteratively performed until the energy function is optimized. In the experiments, more accurate boundaries were obtained, although the proposed method used a simple 3D reconstruction method.

Document scanning is important as a prerequisite stage for analysis and recognition. Recently, a lot of researches about document image acquisition using a camera have been attempted, and the camera can be an alternative input device for document scanning if we can solve some problems such as the low resolution. We use an image registration to overcome the low resolution of a camera. An ordinary image registration method needs a pre-processing such as a camera calibration to reduce distortions on the composite. Therefore the ordinary method has an extra running time. In this paper, we propose a component-based image registration method to concentrate on reducing the distortions and acquiring a seamless image using a PTZ (pan-tilt-zoom) camera without pre-processing. Because we divide the input document image into each component and generate the registration on text components using a text-specific characteristic, this method leads to reduce the object (text) distortions on the composite, and we save the extra running time because this method does not perform the post processing.

The current paper proposes a font classification method for document images that uses non-negative matrix factorization (NMF), that is able to learn part-based representations of objects. The basic idea of the proposed method is based on the fact that the characteristics of each font are derived from parts of individual characters in each font rather than holistic textures. Spatial localities, parts composing of font images, are automatically extracted using NMF and, then, used as features representing each font. Using hierarchical clustering algorithm, these feature sets are generalized for font classification, resulting in the prototype templates construction. In both the prototype construction and font classification, earth mover's distance (EMD) is used as the distance metric, which is more suitable for the NMF feature space than Cosine or Euclidean distance. In the experimental results, the distribution of features and the appropriateness of the features specifying each font are investigated, and the results are compared with a related algorithm: principal component analysis (PCA). The proposed method is expected to improve the performance of optical character recognition (OCR), document indexing and retrieval systems, when such systems adopt a font classifier as a preprocessor.

The current paper presents a novel texture-based method for detecting texts in images. A support vector machine (SVM) is used to analyze the textural properties of texts. No external texture feature extraction module is used, but rather the intensities of the raw pixels that make up the textural pattern are fed directly to the SVM, which works well even in high-dimensional spaces. Next, text regions are identified by applying a continuously adaptive mean shift algorithm (CAMSHIFT) to the results of the texture analysis. The combination of CAMSHIFT and SVMs produces both robust and efficient text detection, as time-consuming texture analyses for less relevant pixels are restricted, leaving only a small part of the input image to be texture-analyzed.

This paper presents a real-time face recognition system. For the system to be real time, no external time-consuming feature extraction method is used, rather the gray-level values of the raw pixels that make up the face pattern are fed directly to the recognizer. In order to absorb the resulting high dimensionality of the input space, support vector machines (SVMs), which are known to work well even in high-dimensional space, are used as the face recognizer. Furthermore, a modified form of polynomial kernel (local correlation kernel) is utilized to take account of prior knowledge about facial structures and is used as the alternative feature extractor. Since SVMs were originally developed for two-class classification, their basic scheme is extended for multiface recognition by adopting one-per-class decomposition. In order to make a final classification from several one-per-class SVM outputs, a neural network (NN) is used as the arbitrator. Experiments with ORL database show a recognition rate of 97.9% and speed of 0.22 seconds per face with 40 classes.

This paper investigates the application of support vector machines (SVMs) in texture classification. Instead of relying on an external feature extractor, the SVM receives the gray-level values of the raw pixels, as SVMs can generalize well even in high-dimensional spaces. Furthermore, it is shown that SVMs can incorporate conventional texture feature extraction methods within their own architecture, while also providing solutions to problems inherent in these methods. One-against-others decomposition is adopted to apply binary SVMs to multitexture classification, plus a neural network is used as an arbitrator to make final classifications from several one-against-others SVM outputs. Experimental results demonstrate the effectiveness of SVMs in texture classification.

A kernel principal component analysis (PCA) was previously proposed as a nonlinear extension of a PCA. The basic idea is to first map the input space into a feature space via nonlinear mapping and then compute the principal components in that feature space. This article adopts the kernel PCA as a mechanism for extracting facial features. Through adopting a polynomial kernel, the principal components can be computed within the space spanned by high-order correlations of input pixels making up a facial image, thereby producing a good performance.

The retrieval of video clips from multimedia databases has been increasingly spotlighted. Texts in videos include useful information for automatic annotation or indexing. Text location is the first step for recognizing the textual information. This paper proposes a neural network-based text location method for news video indexing. Text can be characterized by texture, location, alignment, and font size. The proposed method classifies text pixels and non-text pixels using a network that operates as a set of texture discrimination filters. We find and locate text regions using histogram analysis after removing errors in the classification results. Experimental results show that the proposed method is effective at locating texts.