Two-Stream Convolutional Networks for Dynamic Texture Synthesis

1 code implementation • CVPR 2018 • Matthew Tesfaldet, Marcus A. Brubaker, Konstantinos G. Derpanis

Given an input dynamic texture, statistics of filter responses from the object recognition ConvNet encapsulate the per-frame appearance of the input texture, while statistics of filter responses from the optical flow ConvNet model its dynamics.

Object Recognition Optical Flow Estimation +3