Video summarization is defined as creating a video summary which includes only important scenes in the original video streams. In order to realize automatic video summarization, the significance of each scene needs to be determined. When targeted especially on broadcast sports videos, a play scene, which corresponds to a play, can be considered as a scene unit. The significance of every play scene can generally be determined based on the importance of the play in the game. Furthermore, the following two issues should be considered: 1) what is important depends on each user's preferences, and 2) the summaries should be tailored for media devices that each user has. Considering the above issues, this paper proposes a unified framework for user and device adaptation in summarizing broadcast sports videos. The proposed framework summarizes sports videos by selecting play scenes based on not only the importance of each play itself but also the users' preferences by using the metadata, which describes the semantic content of videos with keywords, and user profiles, which describe users' preference degrees for the keywords. The selected scenes are then presented in a proper way using various types of media such as video, image, or text according to device profiles which describe the device type. We experimentally verified the effectiveness of user adaptation by examining how the generated summaries are changed by different preference degrees and by comparing our results with/without using user profiles. The validity of device adaptation is also evaluated by conducting questionnaires using PCs and mobile phones as the media devices.

This paper aims at analyzing the performance of an anonymous communication system 3-Mode Net with respect to the number of relay nodes required for communication and sender anonymity. As for the number of relay nodes, we give explicit formulas of the probability distribution, the expectation, and the variance. Considering sender anonymity, we quantify the degree of sender anonymity under a situation where some relay nodes collude with each other. The above analyses use random walk theory, a probability generating function, and their properties. From obtained formulas, we show several conditions for avoiding a situation where the number of relay nodes becomes large, and for providing high sender anonymity. Furthermore, we investigate the relationship between the number of relay nodes and sender anonymity, and give a condition for providing a better performance of 3 MN.

With the spread of high-performance sensors and social network services (SNS) and the remarkable advances in machine learning technologies, fake media such as fake videos, spoofed voices, and fake reviews that are generated using high-quality learning data and are very close to the real thing are causing serious social problems. We launched a research project, the Media Clone (MC) project, to protect receivers of replicas of real media called media clones (MCs) skillfully fabricated by means of media processing technologies. Our aim is to achieve a communication system that can defend against MC attacks and help ensure safe and reliable communication. This paper describes the results of research in two of the five themes in the MC project: 1) verification of the capability of generating various types of media clones such as audio, visual, and text derived from fake information and 2) realization of a protection shield for media clones' attacks by recognizing them.

Automatic prosody labeling is the task of automatically annotating prosodic labels such as syllable stresses or break indices into speech corpora. Prosody-labeled corpora are important for speech synthesis and automatic speech understanding. However, the subtleness of physical features makes accurate labeling difficult. Since errors in the prosodic labels can lead to incorrect prosody estimation and unnatural synthetic sound, the accuracy of the labels is a key factor for text-to-speech (TTS) systems. In particular, mora accent labels relevant to pitch are very important for Japanese, since Japanese is a pitch-accent language and Japanese people have a particularly keen sense of pitch accents. However, the determination of the mora accents of Japanese is a more difficult task than English stress detection in a way. This is because the context of words changes the mora accents within the word, which is different from English stress where the stress is normally put at the lexical primary stress of a word. In this paper, we propose a method that can accurately determine the prosodic labels of Japanese using both acoustic and linguistic models. A speaker-independent linguistic model provides mora-level knowledge about the possible correct accentuations in Japanese, and contributes to reduction of the required size of the speaker-dependent speech corpus for training the other stochastic models. Our experiments show the effectiveness of the combination of models.

In this paper, we present a method for extracting the Japanese printed characters from unformatted document images. This research takes into account the multiple general features specific to the Japanese printed characters. In our method, these features are thought of as the constraints for the regions to be extracted within the constraint satisfaction approach. This is achieved by minimizing a constraint function estimating quantitative satisfaction of the features. Our method is applicable to all kinds of the Japanese documents because it is no need of a priori knowledge about the document layout. We have favorable experimental results for the effectiveness of this method.

In this paper, a novel approach is proposed to identify the detailed typeface of Gothic characters in document images. The identification is performed by evaluating two types of typeface models, named the Gs-pattern and the Gd-pattern according to the principle of MDL. The typeface models are generated from the observed character image by using morphology and are viewed as approximating expressions of the observed character. Consequently, this method is unique in that it is free from both character recognition and dictionary lookup.

Our society has been getting more privacy-sensitive. Diverse information is given by users to information and communications technology (ICT) systems such as IC cards benefiting them. The information is stored as so-called big data, and there is concern over privacy violation. Visual information such as images and videos is also considered privacy-sensitive. The growing deployment of surveillance cameras and social network services has caused a privacy problem of information given from various sensors. To protect privacy of subjects presented in visual information, their face or figure is processed by means of pixelization or blurring. As image analysis technologies have made considerable progress, many attempts to automatically process flexible privacy protection have been made since 2000, and utilization of privacy information under some restrictions has been taken into account in recent years. This paper addresses the recent progress of privacy protection for visual information, showing our research projects: PriSurv, Digital Diorama (DD), and Mobile Privacy Protection (MPP). Furthermore, we discuss Harmonized Information Field (HIFI) for appropriate utilization of protected privacy information in a specific area.

Spoofing attacks are one of the biggest concerns for most biometric recognition systems. This will be also the case with silhouette-based gait recognition in the near future. So far, gait recognition has been fortunately out of the scope of spoofing attacks. However, it is becoming a real threat with the rapid growth and spread of deep neural network-based multimedia generation techniques, which will allow attackers to generate a fake video of gait silhouettes resembling a target person's walking motion. We refer to such computer-generated fake silhouettes as gait silhouette clones (GSCs). To deal with the future threat caused by GSCs, in this paper, we propose a supervised method for discriminating GSCs from genuine gait silhouettes (GGSs) that are observed from actual walking people. For training a good discriminator, it is important to collect training datasets of both GGSs and GSCs which do not differ from each other in any aspect other than genuineness. To this end, we propose to generate a training set of GSCs from GGSs by transforming them using multiple autoencoders. The generated GSCs are used together with their original GGSs for training the discriminator. In our experiments, the proposed method achieved the recognition accuracy of up to 94% for several test datasets, which demonstrates the effectiveness and the generality of the proposed method.

The recent popularization of social network services (SNSs), such as YouTube, Dailymotion, and Facebook, enables people to easily publish their personal videos taken with mobile cameras. However, at the same time, such popularity has raised a new problem: video privacy. In such social videos, the privacy of people, i.e., their appearances, must be protected, but naively obscuring all people might spoil the video content. To address this problem, we focus on videographers' capture intentions. In a social video, some persons are usually essential for the video content. They are intentionally captured by the videographers, called intentionally captured persons (ICPs), and the others are accidentally framed-in (non-ICPs). Videos containing the appearances of the non-ICPs might violate their privacy. In this paper, we developed a system called BEPS, which adopts a novel conditional random field (CRF)-based method for ICP detection, as well as a novel approach to obscure non-ICPs and preserve ICPs using background estimation. BEPS reduces the burden of manually obscuring the appearances of the non-ICPs before uploading the video to SNSs. Compared with conventional systems, the following are the main advantages of BEPS: (i) it maintains the video content, and (ii) it is immune to the failure of person detection; false positives in person detection do not violate privacy. Our experimental results successfully validated these two advantages.

This paper describes COCE (COordinative Character Extractor), a method for extracting printed Japanese characters and their character strings from all sorts of document images. COCE is based on a multi-agent system where each agent tries to find a character string and extracts the characters in it. For the adaptability, the agents are allowed to look after arbitrary parts of documents and extract the characters using only the knowledge independent of the layouts. Moreover, the agents check and correct their results sometimes with the help of the other agents. From experimental results, we have verified the effectiveness of our approach.

This paper discusses the role of knowledge in document image understanding from the viewpoints of representation, utilization and acquisition. For the representation of knowledge, we propose two models, a layout model and a content model, which represent knowledge about the layout structure and content of a document, respectively. For the utilization of knowledge, we implement layout analysis and content analysis which utilize a layout model and a content model, respectively. The strategy of hypothesis generation and verification is introduced in order to integrate these two kinds of analysis. For the acquisition of knowledge, we propose a method of incremental acquisition of a layout model from a stream of example documents. From the experimental results of document image understanding and knowledge acquisition using 50 samples of visiting cards, we verified the effectiveness of the proposed method.

Recently, a number of location-based services such as navigation and mobile advertising have been proposed. Such services require real-time user positions. Since a global positioning system (GPS), which is one of the most well-known techniques for real-time positioning, is unsuitable for indoor uses due to unavailability of GPS signals, many indoor positioning systems (IPSs) using WLAN, radio frequency identification tags, and so forth have been proposed. However, most of them suffer from high installation costs. In this paper, we propose a novel IPS for real-time positioning that utilizes a digital audio watermarking technique. The proposed IPS first embeds watermarks into an audio signal to generate watermarked signals, each of which is then emitted from a corresponding speaker installed in a target environment. A user of the proposed IPS receives the watermarked signals with a mobile device equipped with a microphone, and the watermarks are detected in the received signal. For positioning, we model various effects upon watermarks due to propagation in the air, i.e., delays, attenuation, and diffraction. The model enables the proposed IPS to accurately locate the user based on the watermarks detected in the received signal. The proposed IPS can be easily deployed with a low installation cost because the IPS can work with off-the-shelf speakers that have been already installed in most of the indoor environments such as department stores, amusement arcades, and airports. We experimentally evaluate the accuracy of positioning and show that the proposed IPS locates the user in a 6 m by 7.5 m room with root mean squared error of 2.25 m on average. The results also demonstrate the potential capability of real-time positioning with the proposed IPS.

Fake media has been spreading due to remarkable advances in media processing and machine leaning technologies, causing serious problems in society. We are conducting a research project called Media Clone aimed at developing methods for protecting people from fake but skillfully fabricated replicas of real media called media clones. Such media can be created from fake information about a specific person. Our goal is to develop a trusted communication system that can defend against attacks of media clones. This paper describes some research results of the Media Clone project, in particular, various methods for protecting personal information against generating fake information. We focus on 1) fake information generation in the physical world, 2) anonymization and abstraction in the cyber world, and 3) modeling of media clone attacks.