The book "Perceptual Organization and Visual Recognition" by David G. Lowe, published by Kluwer Academic Publishers, explores the field of computer vision and its application in robotics. The author, David G. Lowe, is affiliated with the Courant Institute of Mathematical Sciences at New York University. The book is part of the Kluwer International Series in Engineering and Computer Science, focusing on robotics and vision. The content is structured into several chapters, each addressing different aspects of visual recognition and perceptual organization. Key topics include: 1. **Introduction**: Overview of spatial vision, perceptual organization, and the relationship to psychology. 2. **Previous Research**: Examination of Gestalt psychology, the principle of simplicity, and the role of grouping in computer vision systems. 3. **The Significance of Image Relations**: Discussion on probability, viewpoint invariance, and computational complexity. 4. **The Segmentation of Image Curves**: Techniques for segmenting image curves and selecting significant structures. 5. **The Use of Viewpoint Invariance**: Inferences from image features, recovery of 3D properties, and three-space inference. 6. **Model-based Search and Inference**: Probabilistic search, evidential reasoning, and the verification of interpretations. 7. **The SCERPO Vision System**: Detailed description of the system, including edge detection, perceptual organization, matching, and verification. 8. **Conclusions**: Directions for future development and implications for human vision. The book emphasizes the importance of perceptual organization in visual recognition, particularly in recognizing objects from unknown viewpoints without prior depth information. It also highlights the role of mathematical methods in bringing 3D models into correspondence with 2D image features, enhancing the accuracy and reliability of visual recognition. The author acknowledges the contributions of numerous individuals and institutions throughout the research process.The book "Perceptual Organization and Visual Recognition" by David G. Lowe, published by Kluwer Academic Publishers, explores the field of computer vision and its application in robotics. The author, David G. Lowe, is affiliated with the Courant Institute of Mathematical Sciences at New York University. The book is part of the Kluwer International Series in Engineering and Computer Science, focusing on robotics and vision. The content is structured into several chapters, each addressing different aspects of visual recognition and perceptual organization. Key topics include: 1. **Introduction**: Overview of spatial vision, perceptual organization, and the relationship to psychology. 2. **Previous Research**: Examination of Gestalt psychology, the principle of simplicity, and the role of grouping in computer vision systems. 3. **The Significance of Image Relations**: Discussion on probability, viewpoint invariance, and computational complexity. 4. **The Segmentation of Image Curves**: Techniques for segmenting image curves and selecting significant structures. 5. **The Use of Viewpoint Invariance**: Inferences from image features, recovery of 3D properties, and three-space inference. 6. **Model-based Search and Inference**: Probabilistic search, evidential reasoning, and the verification of interpretations. 7. **The SCERPO Vision System**: Detailed description of the system, including edge detection, perceptual organization, matching, and verification. 8. **Conclusions**: Directions for future development and implications for human vision. The book emphasizes the importance of perceptual organization in visual recognition, particularly in recognizing objects from unknown viewpoints without prior depth information. It also highlights the role of mathematical methods in bringing 3D models into correspondence with 2D image features, enhancing the accuracy and reliability of visual recognition. The author acknowledges the contributions of numerous individuals and institutions throughout the research process.