In this book, Shimon Ullman focuses on the processes of high-level vision that deal with the interpretation and use of what is seen in the image. In particular, he examines two major problems. The first, object recognition and classification, involves recognizing objects despite large variations in appearance caused by changes in viewing position, illumination, occlusion, and object shape. The second, visual cognition, involves the extraction of shape properties and spatial relations in the course of performing visual tasks such as object manipulation, planning movements in the environment, or interpreting graphical material such as diagrams, graphs and maps.The book first takes up object recognition and develops a novel approach to the recognition of three-dimensional objects. It then studies a number of related issues in high-level vision, including object classification, scene segmentation, and visual cognition. Using computational considerations discussed throughout the book, along with psychophysical and biological data, the final chapter proposes a model for the general flow of information in the visual cortex.Understanding vision is a key problem in the brain sciences, human cognition, and artificial intelligence. Because of the interdisciplinary nature of the theories developed in this work, High-Level Vision will be of interest to readers in all three of these fields.

1 Object Recognition
1.1 Shape-Based Recognition 
1.2 What Is Recognition? 
1.3 Why Object Recognition Is Difficult

2 Approaches to Object Recognition
2.1 Invariant Properties and Feature Spaces 
2.2 Parts and Structural Descriptions 
2.3 The Alignment Approach 
2.4 Which Is the Correct Approach?

3 The Alignment of Pictorial Descriptions
3.1 Using Corresponding Features 
3.2 The Use of Multiple Models for 3-D Objects 
3.3 Aligning Pictorial Descriptions 
3.4 Transforming the Image or the Models? 
3.5 Before and After Alignment

4 The Alignment of Smooth Bounding Contours
4.1 The Curvature Method 
4.2 Accuracy of the Curvature Method 
4.3 Empirical Testing

5 Recognition by the Combination of Views
5.1 Modeling Objects by View Combinations 
5.2 Objects with Sharp Edges 
5.3 Using Two Views Only 
5.4 Using a Single View 
5.5 The Use of Depth Values 
5.6 Summary of the Basic Scheme 
5.7 Objects with Smooth Boundaries 
5.8 Recognition by Image Combinations 
5.9 Extensions to the View-Combination Scheme 
5.10 Psychophysical and Physiological Evidence 
5.11 Interim Conclusions: Recognition by Multiple Views

6 Classification
6.1 Classification and Identification 
6.2 The Role of Object Classification 
6.3 Class-based Processing 
6.4 Using Class Prototypes 
6.5 Pictorial Classification 
6.6 Evidence from Psychology and Biology 
6.7 Are Classes in the World or in Our Head? 
6.8 The Organization of Recognition Memory

7 Image and Model Correspondence
7.1 Feature Correspondence 
7.2 Contour Matching 
7.3 Correspondence-less Methods 
7.4 Correspondence Processes in Human Vision 
7.5 Model Construction 
7.6 Compensating for Illumination Changes

8 Segmentation and Saliency
8.1 Is Segmentation Feasible? 
8.2 Bottom-up and Top-down Segmentation 
8.3 Extracting Globally Salient Structures 
8.4 Saliency, Selection, and Completion 
8.5 What Can Bottom-up Segmentation Achieve?

9 Visual Cognition and Visual Routines
9.1 Perceiving "Inside" and "Outside" 
9.2 Spatial Analysis by Visual Routines 
9.3 Conclusions and Open Problems 
9.4 The Elemental Operations 
9.5 The Assembly and Storage of Routines 
9.6 Routines and Recognition

10 Sequence Seeking and Counter Streams: A Model for Visual Cortex
10.1 The Sequence-seeking Scheme 
10.2 Biological Embodiment 
10.3 Summary 
 

Shimon Ullman is Samy and Ruth Cohn Professor of Computer Science at Weizmann Institute of Science, Rehovot, Israel.

배송정보

교환/반품 정보