ICIP 2006, Atlanta, GA

Slide Show

Atlanta Conv. & Vis. Bureau



ICIP 2006 will feature the following tutorials. To register for a tutorial, please click here.

Morning Session, October 8, 2006:

09:00-10:30Part 1
10:50-12:20Part 2
  • Security of Digital Multimedia Content: Solutions With Encryption And Watermarking (Abstract)
    Presenter: Ahmet M. Eskicioglu, Brooklyn College of the City University of New York
  • Image Processing Techniques in Computer-Aided Detection and Diagnosis (Abstract)
    Presenter: Metin N. Gurcan, Ohio State University
  • Real-Time Image and Video Processing: From Research to Reality (Abstract)
    Presenters: Nasser Kehtarnavaz and Mark Gamadia, University of Texas at Dallas

Afternoon Session, October 8, 2006:

14:00-15:30Part 1
15:50-17:20Part 2


  • Security of Digital Multimedia Content: Solutions With Encryption And Watermarking: In recent years, advances in digital technologies have created significant changes in the way we reproduce, distribute and market intellectual property (IP). Digital media can now be exploited by the IP owners to develop new and innovative business models for their products and services. The lowered cost of reproduction, storage and distribution, however, also invites much motivation for large-scale commercial infringement. In a world where piracy is a growing potential threat, the rights of the IP owners can be protected using three complementary weapons: Technology, legislation, and business models. Because of the diversity of IP (ranging from ebooks to songs and movies), no single solution is applicable to the protection of multimedia products in distribution networks.

    End-to-end security is the most critical requirement for the creation of new digital markets where copyrighted multimedia content is a key product. Three major industries have a vital interest in this problem: The motion picture industry, the consumer electronics (CE) industry, and the information technology (IT) industry. This tutorial is an overview of the work done for protecting the content owners' investment in intellectual property. After an introduction to copyright and copyright industries, we examine how the technological, legal, and business solutions help maintain the incentive to supply the lifeblood of the markets.
  • Image Processing Techniques in Computer-Aided Detection and Diagnosis: Computer-aided Detection/Diagnosis (CAD) is truly an interdisciplinary research area. Development of a CAD system requires coordinated efforts of medical professionals, algorithmic and software engineers, and statisticians. Image processing techniques are frequently used in every aspect of the development from initial pre-processing techniques for noise reduction to segmentation of lesions and to registration of tumors. In this tutorial, some exciting CAD problems will be introduced and several commonly used image processing techniques will be reviewed within the context of medical image analysis for CAD. The tutorial will be taught from the perspective of a researcher with an image processing background, who carried out research in this field for over 10 years both in academia and industry.

    Outline of the Tutorial:
    • Definition of computer-aided detection, diagnosis (CAD)
    • Brief history of CAD research
    • Examples of CAD systems in radiology
      • Detection of masses and microcalcifications from mammograms
      • Detection of solid pulmonary nodules from computed tomography images
      • Detection of adenatomous polyps from virtual colonoscopy studies
    • Image processing applications in CAD
      • Pre-processing techniques
      • Enhancement of lesions
      • Segmentation of lesions
      • Registration of lesions
      • Multi-resolution image processing techniques in CAD
    • Future directions
  • Real-Time Image and Video Processing: From Research to Reality: This tutorial will discuss the process of transitioning an algorithm from a research development environment to a real-time constrained, hardware platform. In such a process, one has to deal with processing time constraints and also constraints placed on available processing capability, memory, system size, and power consumption. While the goal of transitioning to a real-time implementation is a practical one, the challenges involved often discourage researchers from pursuing algorithms to completion, leaving them to someone else to examine performance tradeoffs, and to implement practical, real-time versions. It is the aim of this tutorial to provide guidelines so that the burden of transitioning an image or video processing algorithm to its real-time implementation is eased. The audience will be introduced to a wide variety of commonly used tools and strategies, which they can then use in the transition process. Key examples from the literature will be presented to illustrate the application of these strategies in actual systems. The topics of the course parallel those in the new book authored by the presenters entitled Real-Time Image and Video Processing: From Research to Reality.
  • Digital Color Management: Encoding Solutions: The principal objectives of color management are to represent, control, and communicate color within and among color-imaging systems. Over the years, numerous color-management methods have been developed in an attempt to meet these objectives. Many have claimed to do so by providing "device-independent" color. In practice, however, none of these attempts has proven completely successful.

    This course will set forth the basic principles required to understand the successful management of color in imaging systems. A “universal” color-management paradigm will be described which, together with its unique appearance-based color encoding, offers a comprehensive solution to the difficult problem of managing color in today’s complex electronic and hybrid color-imaging systems.

    Course Benefits:
    • Understand the fundamental colorimetric principles underlying electronic, traditional, and hybrid color-imaging systems.
    • Ascertain why images from various types of media differ fundamentally in their basic color properties, and the impact these differences have on digital color management.
    • List and compare the capabilities and limitations in the technologies used in various types of color-managed systems.
    • Describe the properties of a universal color-management paradigm.
    • Recognize how the relationship between colorimetry and color appearance can be handled in color-managed systems.
    • Differentiate the universal paradigm’s appearance-based representation from other color-encoding methods.
    • Explain how the universal paradigm can be translated to practical systems.
  • Scalable Video Coding - Standardization and Beyond: The interrelationship and adaptation between transmission/storage and compression technology is one of the most challenging aspects of digital video systems. It is an old dream that an efficient scalable representation of video may provide flexible multi-dimensional resolution adaptation, to support various network and terminal capabilities and also give better error robustness. Currently, a new scalable standard is developed jointly by the Joint Video Team (JVT) of ISO MPEG and ITU-T VCEG as an extension of the successful H.264/MPEG4-AVC project. Unlike previous solutions, it provides a high degree of flexibility in terms of scalability dimensions (supporting various temporal/spatial resolutions, SNR/fidelity levels and global/local ROI access), while the penalty in compression performance as compared to single-layer coding is acceptable. One key factor which has made this possible is the usage of structures modifying the traditional long-recursion loops in motion-compensated video coding. Examples for this are temporal hierarchies of frames and structures with only partial prediction from enhancement layers. When properly combined with Lagrangian optimization techniques and methods for inter-layer prediction (within same and across resolutions) and entropy coding, excellent compression is achievable.

    The purpose of this tutorial is to give a profound insight into this emerging new technology trends. It will give an update on the present status of standardization, and analyze the gains that can be achieved by this new scalable video coding technology theoretically and practically.
  • Biometrics for Surveillance: Of the many biometric approaches available for surveillance applications, face and gait biometrics are natural due to their noninvasiveness, i. e. acquisition of face image and gait sequence in principle requires no cooperation of the participants. Although face recognition has been actively studied over the past decade, the state-of-the-art face recognition systems yield satisfactory performance under controlled conditions and degrade significantly when confronted with illumination and pose variations, aging, expressions, disguises, etc. Gait-based human recognition, as an emerging biometric, generalizes inadequately across surface type, camera viewing angle, load carrying conditions and even shoe type!

    The goal of this tutorial is to provide a comprehensive review of face and gait based human recognition algorithms with applications in surveillance. Specifically, we will discuss methods for face recognition using video sequences, illumination-invariant still face recognition methods, and face recognition across aging, gait-based human identification using fronto-paralllel and arbitrary views.

    The tutorial is accessible to a wide audience since only basic level of linear algebra, probability, statistics, and image processing is assumed. Graduate students and researchers new to the field can use the tutorial to quickly comprehend the state-of-the-art of unconstrained face and gait recognition. Also the tutorial could serve as a starting point for them to embark on their research on face and gait recognition. Designers of surveillance systems can readily extract useful engineering principles that will come in handy in their work.
  • Compressive Sampling: A New Framework for Imaging: Imaging sensors, hardware, and algorithms are under increasing pressure to accommodate ever larger and higher-dimensional data sets; ever faster capture, sampling, and processing rates; ever lower power consumption; communication over ever more difficult channels; and radically new sensing modalities. Fortunately, over the past few decades, there has been an enormous increase in computational power and data storage capacity, which provides a new angle to tackle these challenges. In this tutorial, we will introduce the theoretical background, new algorithms, and practical applications of "Compressive Sampling" (CS), whose motto is: "Sample smarter, not faster". A series of theoretical and practical results developed over the past two years suggests that the number of measurements (e.g., samples) required to capture a signal, image, or video sequence depends more on its intrinsic information content (its compressibility) than the desired resolution. CS is based on novel measurement techniques (including random projections) and exploits modern optimization algorithms for image extraction and processing. The theory of CS, while still in its developing stages, is far-reaching and draws on subjects as varied as sampling theory, convex optimization, source and channel coding, statistical estimation, uncertainty principles, and high-dimensional geometry. The applications of CS range from the familiar (imaging in medicine and radar, high-speed analog-to-digital conversion, and super-resolution) to truly novel image acquisition and encoding techniques.

For further information, please contact Dr. Stan Reeves.