ShapeSifter: a retrieval system for databases of 3D engineering data

Large amounts of 3D data are being generated daily from increasingly diverse sources. In applications ranging from medical imaging and archival scanning of archaeological artifacts through to engineering and design, valuable geometric information is being recorded and stored. The size of these 3D databases make manual indexing time consuming, ineffective and often simply impracticable. Because of this, researchers have started to develop tools for automatically searching and organising these digital libraries of shape information. The most widespread and valuable form of 3D data are the design models created by commercial manufacturing companies. Engineering companies commonly have tens of thousands of 3D Computer-Aided Design models stored on their computer systems. These models are used to communicate the exact shape and dimensions of components to both customers and subcontracting manufacturers. Consequently these models are of great value and importance to the companies. Currently 3D models (like engineering drawings) are indexed by alpha-numeric 'part numbers' with a format unique to each company. Although this system of indexing works well in the context of ongoing maintenance and development of individual parts, it offers little scope for 'data mining' (i.e. exploration) of a company's inventory of designs. This paper describes an online system which allows collections of 3D models to be searched for components geometrically similar to a user-defined target model