Eiji Kasutani

The MPEG-7 color layout descriptor: a compact image feature description for high-speed image/video segment retrieval

This paper proposes a color feature description for image and video retrieving applications such as personal video recorder, which manages a large amount of data. This descriptor specifies the spatial distribution of colors with a few nonlinear quantized DCT coefficients of grid based average colors. The tradeoff between the number of the coefficients enclosed in the descriptor and retrieval cost is studied. The experimental results show that the descriptor enclosing six for luminance and three for each chrominance coefficient achieves the best trade-off between the storage cost and retrieval efficiency. It requires 6 bits for DC and 5 bits for AC coefficients, and therefore total storage cost is just 63 bits per image. This description, named color layout descriptor, has been accepted as a part of the MPEG-7 final committee draft.

Image Signature Robust to Caption Superimposition for Video Sequence Identification

This paper proposes an image signature robust to caption superimposition for video sequence identification. A new image signature which is a set of local features is developed for a high-speed frame-by-frame matching of video sequences. The signature of a frame is obtained by partitioning the image into blocks and extracting the local feature representing the dominant type of edge direction from each block. The similarity between the signatures is calculated by comparing the edge types of the corresponding blocks, and counting the number of the blocks having the same edge type. A weighting scheme based on the probability of caption superimposition for each block can be applied to the similarity calculation to improve the matching performance. The experimental results of the video sequence identification show that the proposed signature achieves precision of 99.65% and recall of 99.45%, improving both the precision and the recall by more than 30% compared with the conventional signature.

Video material archive system for efficient video editing based on media identification

This paper proposes a new media identification related application; a video material archive system for efficient video editing. This system allows content creators to track the usage history of all the archived materials to support their access to the desired materials. It also provides the citation frequency of each material so that creators can measure its importance. A feature-based media identification method is employed to create the usage history and the citation frequency data. Experimental results show that the use of this identification method makes it possible to create those data fully automatically without forcing any changes to the existing devices

An adaptive feature comparison method for real-time video identification

This paper proposes an adaptive feature comparison method for real-time video identification. The proposed method continuously analyzes the input video signal and compares them with prestored video sequences in feature domain adaptively to the analysis result. The number of features for comparison and thresholds for similarity matching are dynamically controlled to identify the input video correctly in real-time. This parameter control enables high-speed matching by minimizing redundant matching and improves robustness against analog noise. Experimental results show that it achieves identification with no errors from a large video clip database and realtime identification from more than 200,000 video clips of 15 seconds.

On image segmentation for object-based image retrieval

We describe a new approach to image segmentation to improve object-based image retrieval. This method partitions color images automatically into disjointed meaningful regions by integrating the contour-based analysis with the region-based analysis. It evaluates the description length of each region and groups multiple regions to form a larger region to minimize the total description length. Through this process, it can distinguish the object boundaries from the edge of the texture. The system successfully extracts the semantically meaningful objects, which meet the segmentation guideline for object-based image retrieval. We have built a system based on this new approach to image segmentation and applied it to a personal photograph database for evaluation. Retrieval results show usefulness and confirm the effectiveness of the proposed methods.

Multimedia Content Filtering, Browsing, and Matching Using MPEG-7 Compact Color Descriptors

This paper presents two compact color feature descriptors that have been accepted to the working draft of MPEG-7, the ongoing activity to standardize multimedia content descriptions. The need for content description arises from the increase in the number of broadcast channels and the availability of a large number of video content over the Internet. Without such descriptions, the average consumer would be left helpless with enormous amount of data and no viable tool to efficiently access the content. MPEG-7 is standardizing descriptions for audio-visual features such as color, shape, and texture, and as well as the higher level semantic features. Here we present two color descriptors that are very compact in size, namely, the compact color histogram descriptor and the compact color layout descriptor. These two descriptors capture complementary characteristics of the color content of images. The compact sizes of these descriptors result in minimal storage and transmission requirements of the image representation. We demonstrate the efficacy of these descriptors for video content filtering, image matching, sketch-based matching and browsing applications.

Linking TV programs with Internet contents based on video fingerprinting

This paper proposes a system that provides TV/Internet converged services by linking video segments in TV programs with their related Internet contents without forcing any changes to TV contents. This system allows users to access the Internet contents related to the video clips currently played back. A robust video fingerprinting method is employed to link video segments with its related contents. Experimental results show that this system makes it possible to create those links fully automatically with no errors

Mobile Multimedia Library: an MPEG-7 application with camera-equipped mobile phones

This paper proposes a Mobile Multimedia Library (MML), a new application of information retrieval using camera-equipped mobile phones. MML allows users to get information of an unknown object anywhere and anytime. A user simply takes a picture of the object by a mobile phone and sends it directly to an MML server on which the picture is analyzed to identify the object. The MML server returns the identification result to the mobile phone and the user can browse information of the object on the display of the mobile phone. This application employs the k-Nearest Neighbor approach using eight MPEG-7 visual features to identify objects. A prototype system for animal identification has been developed to demonstrate effectiveness of the MML framework. It takes about ten seconds on average for the identification in which 63% and 79% queries were correctly identified within first four and first ten candidates, respectively, out of 229 categories.