Abstract: A computer-implemented technique for dip filtering of seismic data whereby seismic data of a selected attribute in space and time coordinates is displayed in accordance with a 3-D illumination model such that the apparent angle of illumination of the data relative to an apparent viewing position is optimally adjusted to achieve best differentiation and display presentation of seismic dipping events, fault indications, excessive multiple indications, and the like. The method requires a perceptual view of a seismic section as a 3-D surface from which a vector of images can be derived. Thereafter, vector of images surface components are derived and, in accordance with a pre-determined illumination model, i.e. composed as a function of azimuth and elevation relative to a display, the vector of images of the seismic data are transformed to produce enhancing a specific eye view/illumination alignment.