Part Decomposition of 3D Surfaces presents a rigorous exploration of methods for segmenting and understanding complex three-dimensional structures. Originally developed as a doctoral dissertation in 2003, this revised edition preserves the core research while presenting it in a format accessible to a broader technical audience.

At the heart of the work lies a fundamental challenge in computer vision and geometric modeling: how to break a continuous surface into meaningful, interpretable components. This problem sits at the intersection of mathematics, computation, and perception. Effective solutions enable downstream tasks such as object recognition, simulation, and geometric analysis.

The methods developed in this volume focus on identifying intrinsic structure within 3D meshes through algorithmic decomposition. Emphasis is placed on balancing mathematical rigor with practical implementation, offering approaches that remain relevant to modern applications in computer graphics, CAD, and 3D imaging.

Beyond the algorithms themselves, the work reflects a moment in the evolution of computer vision when advances in computation began to make large-scale geometric processing feasible. Many of the underlying ideas continue to inform contemporary research in shape analysis, digital twins, and machine perception.

This edition serves both as a historical record and a technical reference and is intended for researchers, graduate students, and practitioners seeking a deeper understanding of how complex surfaces can be reduced to their essential parts-revealing structure where complexity first appears to dominate.