The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The italic detection and slant rectification is a key step of optical character recognition (OCR). In this paper, a novel method is proposed to detect and rectify italic characters in Chinese advertising images. Based on observations on structures of many characters, the centroid angle is proposed and a statistical study on it is presented. According to the statistical results, the centroid angle...
Automatic behavior recognition is one important task of community security and surveillance system. In this paper, a novel method is proposed for automatic selection of behavior models by iterative learning and abnormality recognition. The method is mainly composed of the following two steps: (1) The models of normal behaviors are automatically selected and trained by combining Dynamic Time Warping...
In this paper, we present a system for Chinese news program management based on cross media video analysis. Audio, caption text and video frames are all important for a person to understand the meaning of the video. Given these facts, we devised a system integrating continuous Chinese speech recognition (ASR), video caption text recognition (VOCR) and object/scene recognition (OR). The news program...
It has always been very difficult to recognize realistic actions from unconstrained videos because there are tremendous variations from camera motion, background clutter, object appearance and so on. In this paper, a Single-Feature Hierarchical Latent Dirichlet Allocation model called SF-HLDA by extending Latent Dirichlet Allocation to the hierarchical one is first proposed for realistic action recognition...
In this paper, we call the pattern classification problem that consists in assigning a category label to a long audio signal based on its semantic content as Generic Audio Document Categorization (GADC). A novel generative model is proposed to describe the generic audio document categories and solve the GADC problem. This model is a four-level hierarchical model in which two latent variables “audio...
In recent decades, more television stations in China are providing sign language news reports to assist the deaf and mute to watch the news programs on TV. As the sign language videos are becoming enormous, how to manage them remains formidable. Our objective is to present a method to manage the sign video resource and implement a contentbased analysis and retrieval system. The contributions are:...
The story-related subject caption (SSC) in broadcast news video expresses the subject of news story, and plays an important role in news story segmentation and news video indexing. We find that a SSC always has a strip background and all the SSCs in one news video have the same style. By taking advantage of these characters, this paper presents an unsupervised approach to detect SSCs in broadcast...
It has been a challenge to locate character in flash due to the variations of character in size, color and style. Besides, multilingual text in flash brings more difficulties to character localization. In this paper, a novel character localization method based on connected component analysis is proposed for locating character in flash. The proposed method first clusters the color to separate the color...
With the booming of online economy, more and more advertisements appear on the network, meanwhile many illegal advertisements emerge. To detect the advertisements automatically, a Web-based advertising content analysis platform is proposed in this paper. This platform consists of the following three parts: web information extraction, advertiser's named entity identification and advertiser's industry...
We consider the image classification problem based on the similarities between images. The choice of the similarity is related to the particular applications, and it could be based on color, texture, bag-of-features, or even more complex kernels. As long as the pair-wise similarity matrix is transformed into a positive semidefinite one, the similarities of images could be treated as kernels. This...
In this paper, we present a revised method to compute the similarity of traditional string edit distance. Given two strings X and Y over a finite alphabet, an edit distance between X and Y can be defined as the minimum weight of transforming X into Y through a sequence of weighted edit operations. Because this method lacks some type of normalization, it would bring some computation errors when the...
This paper presents a system called DCMR. Content-based video searching is a challenging field, and most research focus on the low level features such as color histogram, texture and etc. In this paper, we solve the searching problem by high level features used by hand language recognition. Firstly, we find the face in video frames that has complex background, and then we find the left hand and right...
A fast image in painting method based on hybrid similarity-distance is proposed in this paper. In Criminisi et al.'s work, similarity distance are not reliable enough in many cases and the algorithm performs inefficiently. To solve these problems, we propose a new searching strategy to accelerate the algorithm. In addition, we modify the confidence-updating rule to make more reasonable the distributions...
In this paper, we solve the searching problem by high level features used by hand language recognition. Firstly, we find the face in video frames that has complex background, and then we find the left hand and right hand in specific areas. By computing the hands' length, position, velocity, acceleration, Fourier figure descriptor and etc, we generate the hands' dynamic features. Consequently, we segment...
We present a novel approach to measuring similarity between objects based on matching local “appearance contextual descriptor”. The descriptor has two components: Histogram of Oriented Gradient feature representing local patch appearance and the contextual descriptor capturing not only the spatial distribution of the non-reference patches relative to the reference patch but also the appearance similarities...
In digital image watermark applications, most of watermark algorithms are vulnerable to geometrical affine transformation attacks. In order to resist affine transformation attacks, we propose a correlation coefficient based algorithm, which uses the original image to estimate affine parameters and recover the affine transformed image before watermark detection. Compared with existing affine parameter...
In this paper, we present a hierarchical framework for detecting and localizing object by components. The system is structured with a root detector and several component detectors that are trained to separately find the object and different parts of the object on the first level. On the second level the spatial relations model performs detection by combining the root detector and the component detectors...
In most of the existing shot boundary detection algorithms, the false/miss detection problem caused by motion is very serious. In this paper, firstly, we propose a new spatio-temporal slice called projected spatio-temporal slice (PSTS) that can effectively eliminate disturbance caused by motion. Then we present approaches for detecting camera cuts, fades and dissolves based on motion estimation of...
In recent years, digital watermarking technologies are considered to be important means to protect digital image copyrights. In this paper, we combine a non-invertible watermark scheme with an HVS model adaptive watermark. The application of adaptive watermarks improves visual imperceptibility. However, we find that an adversary is able to apply ambiguity attacks on the non-invertible watermark scheme...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.