These sequences were used for our video interpolation work described in High-quality video view interpolation using a layered representation, C.L. Zitnick, S.B. Kang, M. Uyttendaele, S. Winder, and R. Szeliski, ACM SIGGRAPH and ACM Trans. on Graphics, Los Angeles, CA, Aug. 2004, pp. 600-608. Each sequence is 100 frames long. The camera resolution is 1024x768 and the capture rate is 15fps. The depth data were computed using the stereo technique described in our Siggraph 2004 paper and should not be regarded as ground truth. Each dataset was broken up into five parts due to its large size. Breakdancing Sequence Ballet Sequence

Related datasets