Isidore Rigoutsos Robert Hummel (2 results)

Paperback. Condition: New. Print on Demand. This book explores the creation of a parallel implementation for geometric hashing on the Connection Machine. Geometric hashing is a powerful method for model-based recognition, allowing for the fast, accurate detection of objects in complex scenes by creating a hash table data structure that encodes information about the models in a highly redundant, multiple-viewpoint way. The author provides a scalable parallel algorithm (requiring Mn3 processors), a novel remapping function to achieve uniform distribution of entries over the rectangular hash table, and an innovative technique to fully exploit symmetries due to combinations of basis points, proving that no more than a few tens of probes are needed to achieve accurate recognition. This book is a significant contribution to the field of computer vision, providing valuable insights into the design and implementation of parallel algorithms for geometric hashing. This book is a reproduction of an important historical work, digitally reconstructed using state-of-the-art technology to preserve the original format. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in the book. print-on-demand item.

Paperback. Condition: New. Print on Demand. This book introduces a new method to model-based object recognition using a parallel geometric hashing algorithm for hypercube SIMD architectures. The algorithm executes in parallel, reducing search time, and enabling fast recognition of objects independent of translation, rotation, and scaling. The author provides multiple building-block algorithms used in the approach, along with a novel radix-sort based histogramming algorithm. The book has broad historical context within both computer science and engineering, discussing the evolution of object recognition. Thematic depth includes analysis of efficient parallel implementation, along with examination of the effectiveness of the algorithm, which can locate embedded objects even when they are obscured or misplaced. This book will appeal to students, academics, and practitioners in computer science, image processing, and engineering. Its insights are highly significant within the field of computer vision and pave the way for exciting new developments. This book is a reproduction of an important historical work, digitally reconstructed using state-of-the-art technology to preserve the original format. In rare cases, an imperfection in the original, such as a blemish or missing page, may be replicated in the book. print-on-demand item.