Abstract

Existing optical flflow datasets are limited in size and variability due to the diffificulty of capturing dense ground truth. In this paper, we tackle this problem by tracking pixels through densely sampled space-time volumes recorded with a high-speed video camera. Our model exploits the linearity of small motions and reasons about occlusions from multiple frames. Using our technique, we are able to establish accurate reference flflow fifields outside the laboratory in natural environments. Besides, we show how our predictions can be used to augment the input images with realistic motion blur. We demonstrate the quality of the produced flflow fifields on synthetic and real-world datasets. Finally, we collect a novel challenging optical flflow dataset by applying our technique on data from a high-speed camera and analyze the performance of the state-of-the-art in optical flflow under various levels of motion blur.