ADVANCED VIDEO CODING: PRINCIPLES AND TECHNIQUES, 7
The Content-based Approach To order this title, and for more information, click here
By K.N. Ngan T. Meier D. Chai, The University of Western Australia, Dept. of Electrical and Electronic Engineering, Visual Communications Research Group, Nedlands, Western Australia 6907, Australia
Description In recent years, the paradigm of video coding has shifted from that of a frame-based approach to a content-based approach, particularly
with the finalization of the ISO multimedia coding standard, MPEG-4. MPEG-4 is the emerging standard for the coding of multimedia content.
It defines a syntax for a set of content-based functionalities, namely, content-based interactivity, compression and universal access.
However, it does not specify how the video content is to be generated. To generate the video content, video has to be segmented into
video objects and tracked as they transverse across the video frames. This book addresses the difficult problem of video segmentation,
and the extraction and tracking of video object planes as defined in MPEG-4. It then focuses on the specific issue of face segmentation
and coding as applied to videoconferencing in order to improve the quality of videoconferencing images especially in the facial region.
Modal-based coding is a content-based coding technique used to code synthetic objects that have become an important part of video
content. It results in extremely low bit rates because only the parameters needed to represent the modal are transmitted. Model-based
coding is included to provide background information for the synthetic object coding in MPEG-4. Lastly, MPEG-4, the first coding standard
for multimedia content is described in detail. The topics covered include the coding of audio objects, the coding of natural and synthetic
video objects, and error resilience. Advanced Video Coding is one of the first books on content-based coding and MPEG-4
coding standard. It serves as an excellent information source and reference for both researchers and practicing engineers.
Contents
Preface. Acknowledgements. Image and Video Segmentation. Bayesian Inference and MRF's. MAP estimation. Markov random
fields (MRFs). Numerical approximations. Edge Detection. Gradient operations: Sobel, Prewitt, Frei-Chen. Canny operator. Image Segmentation.
Morphological segmentation. Bayesia segmentation. Motion. Real motion and apparent motion. The optical flow constraint (OFC). Non-parametric
motion field representation. Parametric motion field representation. The occlusion problem. Motion Estimation. Gradient-based methods.
Block-based techniques. Pixel-recursive algoriths. Bayesian approaches. Motion Segmentation. 3-D segmentation. Segmentation based on
motion information only. Spatio-temporal segmentation. Joint motion estimation and segmentation. References.
Face Segmentation.
Face Segmentation Problem. Various Approaches. Shape analysis. Motion analysis. Statistical analysis. Color analysis. Applications. Coding
area of interest with better quality. Content-based representation and MPEG-4. 3D human face model fitting. Image enhancement. Face recognition,
classification and identification. Face tracking. Facial expression study. Multimedia database indexing. Modeling of Human Skin Color.
Color space. Limitations of color segmentation. Skin Color Map Approach. Face segmentation algorithm. Stage one - color segmentation.
Stage two - density regularization. Stage three - luminance regularization. Stage four - geometric correction. Stage five - contour extraction.
Experimental results. References.
Foreground/Background Coding. Introduction. Related works. Foreground and background
regions. Content-based Bit Allocation. Maximum bit transfer. Joint bit assignment. Content-based Rate Control. H.261FB Approach. H.261
Video coding system. Reference model 8. Implementation of the H. 261FB coder. Experimental results. H.263FB Approach. Implementation
of the H.263FB coder. Experimental results. Towards MPEG-4 Video Coding. MPEG-4 coder. Summary. References.
Model-Based Coding.
Introduction. 2-D model-based approaches. 3-D model-based approaches. Applications of 3-D model-based coding. 3-D Human Facial Modeling.
Modeling a person's face. Facial Feature Contours Extraction. Rough contour location finding. Image processing. Features extraction using
active contour models. Features extraction using deformable templates. Nose feature points extraction using geometrical properties. WFM
Fitting and Adaptation. Head model adjustment. Eye model adjustment. Eybrow model adjustment. Mouth model adjustment. Analysis of Facial
Image Sequances. Estimation of head motion parameters. Estimation of facial expression parameters. High precision estimation by iteration.
Synthesis of Facial Image Sequences. Facial structure deformation method. Update of 3-D Facial Model. Update of texture information.
Udate of depth information. Transmission bit rates. References.
VOP Extraction and Tracking. Video Object Plane Extraction
Techniques. Outline of VOP Extraction Algorithms. Version I: Morphological Motion Filtering. Global motion estimation. Object motion
detection using morphological motion filtering. Model initialization. Object tracking using the Hausdorff distance. Model update. VOP
extraction. Results. Version II: Change Detection Masks. Object motion detection using CDM. Model initialization. Model update. Background
filter. Results. References.
MPEG-4 Standard. Introduction. MPEG-4 Development Process. Features of the MPEG-4 Standard
[2]. Coded representation of primitive AVOs. Composition of AVOs. Description, synchronization and delivery of streaming data for AVOs.
Interaction with AVOs. Identification of intellectual property. Technical Description of the MPEG-4 Standard. DMIF. Demultiplexing, synchronization
and buffer management. Syntax description. Coding of Audio Object. Natural sound. Synthesized sound. Coding of Natural Visual Objects.
Video Object Plane (VOP). The encoder. Shape coding. Mation estimation and compensation. Texture coding. Prediction and coding of B-VOPs.
Generalized scalable coding. Sprite coding. Still image texture coding. Coding of Synthetic Objects. Facial animation. Body animation.
2-D animated meshes. Error Resilience. Resynchronization. Data recovery. Error concealment. Modes of operation. Error resilience encoding
tools. References.
Books and book related electronic products are priced in US dollars (USD), euro (EUR), and Great Britain Pounds (GBP). USD prices apply to the Americas and Asia Pacific. EUR prices apply in Europe and the Middle East. GBP prices apply to the UK and all other countries.
Home | Elsevier Sites | Privacy Policy | Terms and Conditions | Feedback | Site Map | A Reed Elsevier Company