In this paper, we review works related to big visual data on the Web in the literature of computer vision and multimedia research regarding the following points: (1) Web image acquisition for construction of visual concept database for image/video recognition, (2) Web image application for visual concept analysis and data-driven computer graphics, and (3) real-world sensing through Web images to detect location-dependent and event-related visual information.

抄録全体を表示

In recent years, many human-tracking techniques that use images, lasers, and RFID processing have been proposed for analyzing pedestrian trajectories. In normal situations, such research can be used in marketing support and information provision and safety management linked to pedestrian flow. In disaster situations, it can be used to aid the formulation of disaster response plans and evacuation guidance and to measure the state of damage. In this paper, we extract pedestrian flow using stereo cameras and propose a method for comparing and visualizing pedestrian flow measured over an extended period of time. We confirm the effectiveness of the proposed method from the results of an experiment in which we analyze pedestrian flow measured using a stereo camera over a long period of time at an office complex near Akihabara Station. Recently, people obtain a large amount of information via SNS (social networking service) and this information influences human behavior. Lastly, we clarify this influence by visualizing the flow of SNS information and pedestrians.

抄録全体を表示

This paper proposes a sensing-based real-world content called Digital Diorama, which is a three-dimensional miniature model of the dynamically changing real world created from the real-time data continuously published over the Internet by stationary cameras distributed in public spaces. Digital Diorama is designed to preserve both the privacy of the monitored persons and the intelligibility of the content by superimposing the latest background images and human icons each of which represents a monitored person in the three-dimensional model. By using the data continuously published from 10 stationary cameras installed on one floor of a shopping mall, our prototype Digital Diorama browser was able to construct and dynamically update Digital Diorama in approximately 1 fps. The results of subjective evaluations indicated that utilizing appropriate human icons can improve the intelligibility of the content while preserving the privacy of the monitored persons.

抄録全体を表示

Segmentation of document images into text or drawings is an important process, which is often related to binarization of a document image to perform character recognition and document analysis. This process is easier to do using a document image with a uniform background and illuminated under well-conditioned lighting. However, when a document image is very unevenly shaded, binarization becomes very difficult or even impossible. An effective method is to remove the shading prior to binarization. In this paper, we propose a novel method for estimating an unevenly shaded surface in an image obtained under poor illumination. A one-dimensional Gaussian kernel model is applied in both the horizontal and vertical directions to estimate the background surface, allowing uneven shading to be removed from the document image. Thereafter, the image can easily be binarized. Results of experiments conducted on many document images demonstrate that our method yields better results than other methods.

抄録全体を表示

In this paper, we propose a novel packet transmission scheme to improve retransmission efficiency in a bidirectional field pick-up unit (FPU). The proposed scheme's effects on throughput and average delay time are confirmed through the evaluation of computer simulations.
FPU systems, such as television (TV) rebroadcast systems, transmit recorded movies or live footage from terminals to satellite stations by remote broadcast. Because high-definition broadcasting requires reliable transmission systems to carry high-resolution images and sound quality, the development of a bidirectional FPU with a packet exchanging process has become critical in recent years.
To improve transmission efficiency, we propose a packet-based bidirectional transmission system that uses both an efficient retransmission control scheme and a burst transmission scheme. Computer simulation results demonstrate that throughput of the proposed scheme is up to approximately five times greater than that of a conventional FPU, while its transmission average delay is approximately 80% less.

抄録全体を表示

This paper proposes a method for identifying the video that retains the most information from its parent video. Since the parent video is often unavailable, the proposed method estimates its content through the collaborative use of the available video signals that are edited copies of the parent video. By reducing the difference between the video signals of the edited videos, the proposed method then enables the use of conventional no-reference video quality assessment algorithms. Since editing a video requires recompressing it, and since quality assessment algorithms can detect signs of recompression, the proposed method can identify the edited video that retains the most information from the parent video. The effectiveness of the proposed method is verified by subjective experiments over artifical and real-world data sets that include a total of over 400 videos.

抄録全体を表示