Benchmarking CAD search techniques

Abstract

While benchmark datasets have been proposed for testing computer vision and 3D shape retrieval algorithms, no such datasets have yet been put forward to assess the relevance of these techniques for engineering problems. This paper presents several distinctive benchmark datasets for evaluating techniques for automated classification and retrieval of CAD objects. These datasets include (1) a dataset of CAD primitives (such as those common in constructive solid geometry modeling); (2) two datasets consisting of classes generated by minor topological variation; (3) two datasets of industrial CAD models classified based on object function and manufacturing process, respectively; (4) and a dataset of LEGO c models from the Mindstorms c robotics kits. Each model in the datasets is available in three formats -ACIS SAT, ISO STEP, and as a VRML mesh (some models are available under several different fidelity settings). These are all available through the National Design Repository.Using these datasets, we present comprehensive empirical results for nine (9) different shape and solid model matching and retrieval techniques. These experiments show, as expected, that the quality of precision-recall performance can significantly vary on different datasets. These experiments reveal that for certain object classes and classifications, such as those based on manufacturing processes, all existing techniques perform poorly. This study reveals the strengths and weaknesses of existing research in these areas, introduces open challenge problems, and provides meaningful datasets and metrics against which the success of current and future work can be measured.

A framework for similarity recognition of CAD models

“…Among the commonly used evaluation methods for shape benchmarking are Princeton Shape Benchmark (PSB) [35] and National Design Repository (NDR) belong to Drexel University [36]. In the current project, one of the most extensive dataset for shape benchmarking, Engineering Shape Benchmark (ESB) developed in Purdue University has been applied based on the advantages discussed by Jayanti et al [37] PSB includes 867 models in 3D form in three main so called super-class comprising Flat-Thin wall components (107 models), Rectangular-cubic prism (281 models) and Solids of revolution (479 models).…”

KBRE: a proposition of a reverse engineering process by a KBE system

“…Based on benchmark and multimedia retrieval CAD models such as PSB [24], AIM@Shape [25] and [26], KBRE suggests to classify a family of mechanical components. Thus, the classification groups a set of mechanical parts which define a family of mechanical components such as: the piston, the housings, the crankshaft etc.…”

Shape alignment and shape orientation analysis-based 3D shape retrieval system

“…The symmetry transform is useful for shape matching. Bespalov et al [24] presented several distinctive benchmark datasets for evaluating techniques for automated classification and retrieval of CAD objects. Laga et al [25] and Liu et al [26] suggested using discrete spherical wavelets or continuous spherical wavelets to analyze the spherical functions defined by the sampling of the distances between surface and the center of mass of an object.…”