Window of visibility: a psychophysical theory of fidelity in time-sampled visual motion displays

Many visual displays, such as movies and television, rely on sampling in the time domain. We derive the spatiotemporal-frequency spectra for some simple moving images and illustrate how these spectra are altered by sampling in the time domain. We construct a simple model of the human perceiver that predicts the critical sample rate required to render sampled and continuous moving images indistinguishable. The rate is shown to depend on the spatial and the temporal acuity of the observer and on the velocity and spatial-frequency content of the image. Several predictions of this model are tested and confirmed. The model is offered as an explanation of many of the phenomena known as apparent motion. Finally, the implications of the model for computer-generated imagery are discussed. considerable applied as well as theoretical interest with the advent of computer-generated displays. The applied ques- tion is: How often must we present a new view for the stroboscopic display to simulate smooth motion faithfully? The theoretical question may be stated: How can a se- quence of stationary images simulate smooth motion, and why is this particular strobe rate required? Previous attempts to answer these questions have suf- fered in part from lack of an objective measure of how well the stroboscopic display simulates a continuous display. The strictest possible criterion for fidelity is considered here: the ability of a human observer to discriminate visual- ly, by whatever means, between stroboscopic and continuous displays. This permits us to determine the conditions un- der which stroboscopic and continuous motion are visually identical. The perceptual identity of continuous and stro- boscopic displays is then explained in terms of the known spatial and temporal properties of the human visual system. This explanation could take either of two forms. We could examine the stimuli and visual mechanisms in terms of their representation in space and time or in terms of spatial and temporal frequency. Although the two explanations are equivalent, the explanation is simpler in the frequency domain. Fahle and Poggio 7 have applied a similar frequen-