A large class of the proximity effect correction schemes being currently used employ so-called control points in their implementation. Control points are the pixel locations in a circuit pattern at which energy received, referred to as exposure, is estimated. Depending on the exposure estimates at control points, the dose to be given to each feature (or subfeature) or the amount of shape modification for each feature is determined. Many proximity correction schemes including the first version of our scheme (PYRAMID 1.0) adopt a fixed selection of control points, which may not be optimal especially for high density, finely feature circuit patterns. Selection of control points greatly affects the accuracy and speed of correction. We have designed a simple but effective method for adaptively selecting control points: In this article we describe the scheme for adaptive control point selection with simulation results showing the successful correction of high-density, fine-feature circuits in less time.