Stacking, either by itself or as a part of depth migration, is usually used for noise suppression in teleseismic receiver function (RF) images. However, stacking is neither the only signal enhancement method available, nor is it the most efficient in the environment of receiver-side source-generated noise typical for RF imaging. We generalize prestack depth migration methodology by introducing numerous signal-enhancement schemes in place of final summation. The method operates in full 3D, incorporates most of the existing imaging techniques, and suggests a generalized framework of RF depth imaging. We present four applications of this technique using the data from the teleseismic Continental Dynamics–Rocky Mountains teleseismic experiment: (1) building common-image gathers to assess depth focusing of RF images, (2) imaging using median and (3) coherency filters for noise suppression, and (4) generalized 3D common conversion point stacking. The results suggest that with the limited volumes and quality of the existing RF datasets, adaptive filters could be superior to record summation used in conventional depth migration.