Youngchoon Park

Counting People in Groups

Cameras are becoming a common tool for automated vision purposes due to their low cost. In an era of growing security concerns, camera surveillance systems have become not only important but also necessary. Algorithms for several tasks such as detecting abandoned objects and tracking people have already been successfully developed. While tracking people is relatively easy, counting people in groups is much more challenging. The mutual occlusions between people in a group make it difficult to provide an exact count. The aim of this work is to present a method of estimating the number of people in group scenarios. Several considerations for counting people are illustrated in this paper, and experimental results of the method are described and discussed.

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.

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.

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.

A comprehensive curriculum for IT education and workforce development: an engineering approach

Noting the shortage of IT professionals nationally [1], we propose a comprehensive curriculum that supports a variety of programs geared to all ages from early school years to retirement and beyond. Current IT workforce development efforts are limited to training, and have not as yet focused on education and professional development. Largely, this is due to a lack of a science underpinning for IT related curricula. Without such a unified science component, a structured organization of information related concepts cannot be derived.Our proposal includes the development of a number of programs addressing the needs of a variety of learners ranging from elementary school through college and beyond. Seven programs, each with a specific emphasis for various groups, are being developed. Such essential issues as industrial-academic liaisons, workforce (re)training, promotional and awareness programs, teacher training, and IT professional role redefinition, are integral pieces of this project. All developments will be firmly founded on the scientific framework of information science and engineering [2].This work is supported by NSF grant DUE-9950168.

Conceptualization and ontology : Tools for efficient storage and retrieval of semantic visual information

Techniques for content=based image or video retrieval are not mature enough to recognize visual semantic completely. Retrieval based on color, size, texture and shape are within the state of the art. Our experiments on human factors in visual information query and retrieval show that visual information retrieval based on the semantic understanding of visual objects and content are more demanding rather than visual appearance based retrieval. Therefore, it is necessary to use captions or text annotations to photos or videos in content access of visual data. In this paper, human factors in text and image searching are carefully investigated. Based on the resulting human factors, a framework for integrated querying of visual information and textual concept is presented. The framework includes ontology- based semantic query expansion through query term rewriting and database navigation within a conceptual hierarchy within multi modal querying environments. To allow similarity based concept retrieval, a new conceptual similarity distance measure between two conceptual entities in a given conceptual space is proposed. The dissimilarity metric is a minimum weighted path length in the corresponding conceptual tree.

Perspective: A Standards-Based System for Manufacturing Information Integration

This paper presents an infrastructure and a prototype system for a manufacturing information system, which is distributed its nature and is able to store, index, manage, retrieve and present business data, inventory data, and manufacturing processes data. The system works with all kinds of information, such as continuous (i.e., stream oriented) data, production (e.g., decision support) data, legacy data, and multimedia data (say, drawings, pictures, audio signals, voice annotations, and video streams). A key criterion is support for content-based information retrieval across all application areas. The main objective is to provide support for automated information transactions. The prototype of our architecture uses JAVA, STEP (ISO 10303) standard, the Internet, and CORBA. A fully functional system, called Perspective, for retrieval of part and manufacturing process has been designed and implemented in a distributed environment. Among other capabilities, the system can retrieve a set of parts or manufacturing processes based on similarity to some desired criteria.

Dictionary learning based object detection and counting in traffic scenes

The objective of object recognition algorithms in computer vision is to quantify the presence or absence of a certain class of objects, for e.g.: bicycles, cars, people, etc. which is highly useful in traffic estimation applications. Sparse signal models and dictionary learning techniques can be utilized to not only classify images as belonging to one class or another, but also to detect the case when two or more of these classes co-occur with the help of augmented dictionaries. We present results comparing the classification accuracy when different image classes occur together. Practical scenarios where such an approach can be applied include forms of intrusion detection i.e., where an object of class B should not co-occur with objects of class A. An example is when there are bicyclists riding on prohibited sidewalks, or a person is trespassing a hazardous area. Mixed class detection in terms of determining semantic content can be performed in a global manner on downscaled versions of images or thumbnails. However to accurately classify an image as belonging to one class or the other, we resort to higher resolution images and localized content examination. With the help of blob tracking we can use this classification method to count objects in traffic videos. The method of feature extraction illustrated in this paper is highly suited to images obtained in practical cases, which are usually of poor quality and lack enough texture for the popular gradient based methods to produce adequate feature points. We demonstrate that by training different types of dictionaries appropriately, we can perform various tasks required for traffic monitoring.

Concept-Based Visual Information Management with Large Lexical Corpus

Most users want to find visual information based on the semantics of visual contents such as a name of person, semantic relations, an action happening in a scene, ...etc. However, techniques for content-based image or video retrieval are not mature enough to recognize visual semantic completely, whereas retrieval based on color, size, texture and shape are within the state of the art. Therefore, smart ways to manage textual annotation is visual information retrieval are necessary. In this paper, a framework for integration of textual and visual content searching mechanism is presented. The proposed framewrok includes ontology-based semantic query processing through efficient semantic similarity measurement. A new conceptual similarity distance measure between two conceptual entities in a large taxonomy structure is proposed and its efficiency is demonstrated. With the proposed method, an information retrieval system can benefit such as (1) reduction of the number of trial-and-errors to find correct keywords, (2) Improvement of precision rates by eliminating the semantic heterogeneity in description, and (3) Improvement of recall rates through precise modeling of concepts and their relations.

Falcon: Scaling Up Hands-Off Crowdsourced Entity Matching to Build Cloud Services

Many works have applied crowdsourcing to entity matching (EM). While promising, these approaches are limited in that they often require a developer to be in the loop. As such, it is difficult for an organization to deploy multiple crowdsourced EM solutions, because there are simply not enough developers. To address this problem, a recent work has proposed Corleone, a solution that crowdsources the entire EM workflow, requiring no developers. While promising, Corleone is severely limited in that it does not scale to large tables. We propose Falcon, a solution that scales up the hands-off crowdsourced EM approach of Corleone, using RDBMS-style query execution and optimization over a Hadoop cluster. Specifically, we define a set of operators and develop efficient implementations. We translate a hands-off crowdsourced EM workflow into a plan consisting of these operators, optimize, then execute the plan. These plans involve both machine and crowd activities, giving rise to novel optimization techniques such as using crowd time to mask machine time. Extensive experiments show that Falcon can scale up to tables of millions of tuples, thus providing a practical solution for hands-off crowdsourced EM, to build cloud-based EM services.