Anna Leschanowsky: Content-Adaptive Readout of Pixels to Reconstruct Spatio-Temporal High Resolution Videos