Feature indexing techniques are promising for object recognition because of their ability to eliminate many feature set matches from consideration without much computation. This work exploits another property of such techniques. They have inherently parallel structure and connectionist network formulations are easy to develop. Once indexing has been performed, a voting scheme such as geometric hashing [Lamdan et al. 1990] can be used to generate object hypotheses in parallel. We give a framework for the connectionist implementation for such indexing and recognition techniques. With sufficient processing elements, recognition can be performed in a small number of time steps. The number of processing elements necessary to achieve peak performance and the fan-in/fan-out required for the processing elements is determined. These techniques have been simulated on a conventional architecture with good results.