Forouzan Golshani

Motion recovery for video content classification

Like other types of digital information, video sequences must be classified based on the semantics of their contents. A more-precise and completer extraction of semantic information will result in a more-effective classification. The most-discernible difference between still images and moving pictures stems from movements and variations. Thus, to go from the realm of still-image repositories to video databases, we must be able to deal with motion. Particularly, we need the ability to classify objects appearing in a video sequence based on their characteristics and features such as shape or color, as well as their movements. By describing the movements that we derive from the process of motion analysis, we introduce a dual hierarchy consisting of spatial and temporal parts for video sequence representation. This gives us the flexibility to examine arbitrary sequences of frames at various levels of abstraction and to retrieve the associated temporal information (say, object trajectories) in addition to the spatial representation. Our algorithm for motion detection uses the motion compensation component of the MPEG video-encoding scheme and then computes trajectories for objects of interest. The specification of a language for retrieval of video based on the spatial as well as motion characteristics is presented.

Rx for semantic video database retrieval

The most prominent difference between still images and moving pictures stems from movements and variations. Thus to go from the realm of still image repositories to video databases, we must be able to deal with motion. Particularly, we need the ability to classify objects appearing in a video sequence based on the movements of each object, as well as their other characteristics and features such as shape or color. By describing motion derived from motion analysis, we introduce a dual hierarchy consisting of spatial and temporal parts for video sequence representation. This gives us the flexibility to examine arbitrary frames at various levels of abstraction, and to retrieve the associated temporal information (say, object trajectories) in addition to the spatial representation. Our algorithm for motion detection uses the motion compensation component of the MPEG video encoding scheme. The algorithm then computes trajectories for objects of interest. The specification of a language for retrieval of video based on the spatial as well as motion characteristics is presented.

Delay-jitter control in multimedia applications

The growing needs of multimedia communications are leading to new developments in providing real-time communication with guarantees. Several extensions have been proposed for different layers of the Open Systems Interconnection Reference Model to accomodate these needs. In this paper, we study methods for guaranteeing delay jitter bounds for high-speed networks in the network and application layers of this model. The method proposed for the network layer provides distributed jitter control. The method proposed for the application layer allows the destination application to control delay jitter. We use a simulation to compare the effects on delay jitter in each method for various scenarios, such as constant bit rate, cross traffic, and bursty data. In addition, the buffer space requirements for accommodating real-time channels are monitored at each node in the network.

A Multimedia Information Repository for Cross Cultural Dance Studies

Multimedia technologies provide effective means for studying the evolution of dance across time and space. The study may be at the micro level which analyzes the development of an individuals performance and the movements of the dancer(s) in 3D space and over the length of the dance. However, at the macro level, diffusion of dance throughout the world over a span of time may be investigated in order to trace particular dance repertoires that may have traveled across various cultures and traditions. Although clearly different with respect to the expected objectives, both micro level analysis and macro level analysis require detailed comparison of patterns on the basis of certain characteristics that are deemed significant for a given dance. These characteristics are diverse in nature and may include such parameters as design formations, use of space (including level, direction, etc.), dynamics, paraphernalia (e.g., swords, sticks, etc.), sound, and color.We present the design of a multimedia information system with two complimentary aims. The first is to automate, to the greatest degree possible, the process of comparison and analysis of dance and human movement. Much of the information about dance exists in the form of video, images, audio and written commentaries, all collected into a digital library. As dance related materials are added, a wide variety of routines are needed to extract the necessary low level features from the multimedia objects. These low level features are then interpreted to human understandable features and patterns, which will be used for analysis by specialists. The second aim is to bring artists and technologists closer in a meaningful way.

Retrieval and delivery of information in multimedia database systems

Abstract EVA is a multimedia database system capable of storage, retrieval, management, analysis and delivery of objects of various media types, including text, audio, images and moving pictures. The interface language deals with the temporal and spatial aspects of multimedia information retrieval and delivery, in addition to the usual capabilities provided by the ordinary database languages. EVA has five classes of operations, namely: operations for querying and updating (i.e. editing) the multimedia information, operations for screen management, temporal operators, operators for specifying rules and constraints, and aggregation (computational) operators. EVA, an extension of the query language Varqa, is a functional language whose notation is based on that of conventional set theory. It is formally defined in an algebraic framework. EVA is object oriented and supports objects, object classes, attributes and methods of objects, and relationships between objects. The current implementation of EVA runs on several different platforms.

Stereoscopic imaging: a real-time, in depth look

There is a general push for biometric-based solutions to replace keys, ID cards, passwords and PINs. Facial recognition, as one of the computational biometrics technologies, has received renewed attention and publicity lately, but for its inaccurate results. One major reason for the inaccuracy is the fact that, generally, facial recognition tools are rooted in 2D imaging methods which are limited to front-profile 2D photographs with a maximum divergence of 20 degrees. 3D facial imaging technology eliminates much of the nagging problems, but the benefits come with the added cost of processing time, especially in the case of stereoscopic imaging. The requirement of timely processing is particularly important in access control applications. The distributed algorithm we propose represents a novel step in solving the 3D imaging problem using power processing. The algorithm enables cameras, fitted with special boards, to generate 3D images in less time than with existing methods. The algorithm exploits the well-known properties/constraints from the stereovision field. Thus, it is very reliable. The obvious impact areas for this work are the capture, display and transmission of stereoscopic images. However, other areas, such as stereoscopic HDTV, can benefit from a faster technique.

The role of color in content-based image retrieval

Human color perception is subjective. In addition to RGB or HSV values representing color content, psychological factors, circumstantial factors, environmental factors, and physiological factors play an important role in encapsulating the color content. Typically, only the RGB or HSV values are used in indexing and retrieval of color images. In this paper, we demonstrate the superior retrieval performance of techniques which employ all of the above factors in color retrieval. We first present the variety of factors involved in human color perception and also an evaluation of the existing color indexing methods. Several interesting problems including comparison of images in the color-perceptual domain and retrieval by color affection are illustrated as potential novel image query mechanisms.

Real-time transmission of multilayer video over ATM networks

With todays network technology, our expectations for high performance networked multimedia applications are increasing. However, no complete solution as yet exists for providing integrated services to networked multimedia systems. The difficulty lies in the nature of the multimedia data, basically video and audio, which is difficult to store and transmit. The coded video is sensitive to delay, delay variation, bit errors and losses. Our research focuses on the transmission of multilayer real-time video traffic over ATM (Asynchronous Transfer Mode) networks. This issue covers several aspects of ATM and multimedia networking such as video application layer, ATM adaptation layers, traffic and buffer management. The main assumption in this research is that the real-time video traffic is content based coded and consists of several video objects. A single video source with various video objects introduces the need for multiple priorities. We propose a multipriority traffic management model for ATM networks that includes a new priority cell mapping algorithm, a new priority based bandwidth allocation approach, a new adaptation layer definition, and a new multipriority buffer access control algorithm. Using multipriority traffic will complicate the traffic management, the buffer management and the work of the ATM adaptation layers. On the other hand, by allowing the network to have several priority layers during a connection, the overall quality for real-time video will be increased. To our knowledge, no method exists for managing a single connection for multipriority video traffic over ATM networks. Our model is new and addresses the essential transmission requirements of near future multipriority video coding systems over broadband networks.

A Language for Content-Based Video Retrieval

We present an effective technique for automatic extraction, representation, and classification of digital video, and a visual language for formulation of queries to access the semantic information contained in digital video. We have devised an algorithm that extracts motion information from a video sequence. This algorithm provides a low-cost extension to the motion compensation component of the MPEG compression algorithm. In this paper, we present a visual language called VEVA for querying multimedia information in general, and video semantic information in particular. Unlike many other proposals that concentrate on browsing the data, VEVA offers a complete set of capabilities for specifying relationships between the image components and formulating queries that search for objects, their motions and their other associated characteristics. VEVA has been shown to be very expressive in this context mainly due to the fact that many types of multimedia information are inherently visual in nature.

ImageRoadMap: A New Content-based Image Retrieval System

We introduce a new content-based image retrieval system, named ImageRoadMap, for retrieval by visual information. ImageRoadMap provides both computer vision capabilities and database management capabilities. We describe the architectural design of the system and its six main components: Image Processing Object, Image Database Object, Domain Management Object, Feature Extraction Object Visual Query Object, and Data Retrieval and Indexing Object. These objects are independent of one another and may be replaced by objects with equivalent or enhanced features.