Synopses & Reviews
Exploration of Visual Data presents latest research efforts in the area of content-based exploration of image and video data. The main objective is to bridge the semantic gap between high-level concepts in the human mind and low-level features extractable by the machines. The two key issues emphasized are "content-awareness" and "user-in-the-loop". The authors provide a comprehensive review on algorithms for visual feature extraction based on color, texture, shape, and structure, and techniques for incorporating such information to aid browsing, exploration, search, and streaming of image and video data. They also discuss issues related to the mixed use of textual and low-level visual features to facilitate more effective access of multimedia data. To bridge the semantic gap, significant recent research efforts have also been put on learning during user interactions, which is also known as "relevance feedback". The difficulty and challenge also come from the personalized information need of each user and a small amount of feedbacks the machine could obtain through real-time user interaction. The authors present and discuss several recently proposed classification and learning techniques that are specifically designed for this problem, with kernel- and boosting-based approaches for nonlinear extensions. Exploration of Visual Data provides state-of-the-art materials on the topics of content-based description of visual data, content-based low-bitrate video streaming, and latest asymmetric and nonlinear relevance feedback algorithms, which to date are unpublished. Exploration of Visual Data will be of interest to researchers, practitioners, and graduate-level students in the areas of multimedia information systems, multimedia databases, computer vision, machine learning.
Table of Contents
1: Introduction. 1.1. Challenges. 1.2. Research Scope. 1.3. State-of-the-Art. 1.4. Outline of Book. 2: Overview Of Visual Information Representation. 2.1. Color. 2.2. Texture. 2.3. Shape. 2.4. Spatial Layout. 2.5. Interest Points. 2.6. Image Segmentation. 2.7. Summary. 3: Edge-based Structural Features. 3.1. Visual Feature Representation. 3.2. Edge-Based Structural Features. 3.3. Experiments and Analysis. 4: Probabilistic Local Structure Models. 4.1. Introduction. 4.2. The Proposed Modeling Scheme. 4.3. Implementation Issues. 4.4. Experiments and Discussion. 4.5. Summary and Discussion. 5: Constructing Table-of-Content for Videos. 5.1. Introduction. 5.2. Related Work. 5.3. The Proposed Approach. 5.4. Determination of the Parameters. 5.5. Experimental Results. 5.6. Conclusions. 6: Nonlinearly Sampled Video Streaming. 6.1. Introduction. 6.2. Problem Statement. 6.3. Frame Saliency Scoring. 6.4. Scenario and Assumptions. 6.5. Minimum Buffer Formulation. 6.6. Limited-Buffer Formulation. 6.7. Extensions and Analysis. 6.8. Experimental Evaluation. 6.9. Discussion. 7: Relevance Feedback for Visual Data Retrieval. 7.1. The Need for User-in-the-Loop. 7.2. Problem Statement. 7.3. Overview of Existing Techniques. 7.4. Learning from Positive Feedbacks. 7.5. Adding Negative Feedbacks: Discriminant Analysis? 7.6. Biased Discriminant Analysis. 7.7. Nonlinear Extensions Using Kernel and Boosting. 7.8. Comparisons and Analysis. 7.9. Relevance Feedback on Image Tiles. 8: Toward Unification of Keywords and Low-Level Contents. 8.1. Introduction. 8.2. Joint Querying and Relevance Feedback. 8.3. Learning Semantic Relations between Keywords. 8.4. Discussion. 9: Future Research Directions. 9.1. Low-level and intermediate-level visual descriptors. 9.2. Learning from user interactions. 9.3. Unsupervised detection of patterns/events. 9.4. Domain-specific applications. References. Index.