Two-Stream Convolutional Networks for Dynamic Texture Synthesis
This work addresses the problem of synthesizing realistic dynamic textures for applications in computer graphics and vision, representing an incremental improvement by combining existing methods for style transfer.
The paper tackles dynamic texture synthesis by introducing a two-stream model using pre-trained ConvNets for appearance and dynamics, generating novel textures that match input quality and temporal evolution, with evaluation via a user study.
We introduce a two-stream model for dynamic texture synthesis. Our model is based on pre-trained convolutional networks (ConvNets) that target two independent tasks: (i) object recognition, and (ii) optical flow prediction. 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. To generate a novel texture, a randomly initialized input sequence is optimized to match the feature statistics from each stream of an example texture. Inspired by recent work on image style transfer and enabled by the two-stream model, we also apply the synthesis approach to combine the texture appearance from one texture with the dynamics of another to generate entirely novel dynamic textures. We show that our approach generates novel, high quality samples that match both the framewise appearance and temporal evolution of input texture. Finally, we quantitatively evaluate our texture synthesis approach with a thorough user study.