Feature recognition is a discipline focusing on the designand implementation of algorithms for detecting manufacturinginformation such as holes, slots, etc. in a solid model. Automatedfeature recognition has been an active research area insolid modeling for many years, and is considered to be a criticalcomponent for CAD/CAM integration. This paper gives foundationsfor understanding the state of the art in feature recognitionresearch. It also establishes the context for a special panel sessionat the 17th ASME International Computers in Engineering Conference.Four papers from different research groups participatedin a benchmarking exercise on feature recognition; each wrotea paper on the results of executing their feature recognition systemson a collected set of parts. This panel session should serveas a catalyst for the feature recognition community to collaboratein establishing standard test parts and performance measures and in identifying and resolving research issues.