Overview of the Eulerian video magnification framework. The system first decomposes the input video sequence into different
spatial frequency bands, and applies the same temporal filter to all bands. The filtered spatial bands are then amplified by a given factor α,
added back to the original signal, and collapsed to generate the output video. The choice of temporal filter and amplification factors can be
tuned to support different applications. For example, we use the system to reveal unseen motions of a Digital SLR camera, caused by the
flipping mirror during a photo burst (camera; full sequences are available in the supplemental video).