A novel object recognition and localization algorithm was developed for the application of bin-picking parts in the industrial environment. The algorithm is capable of quickly recognizing and estimating the pose of the objects in an unorganized/cluttered scene. Based on the oriented point-pair feature vector, the technique matches points in the scene to points on the surface of the original model via an efficient voting process. Groups of features about a point in the scene are used to find probable matching model points in a precompiled database. Sets of candidate models and scene point-pair matches are created and filtered based on a geometric consistency constraint.