Textons can be learned from motion sequences, which provide more information than single image [1]. Figure 1 reports an experiment that various snow flakes can be tracked and their shape learned over the motion sequence. We call the texton with motion patterns as "moton". For more details, please visit Yizhou Wang's web.

Figure 1. The texton templates for the snow pattern learned from a movie sequence. (Zhu, Guo, Wu and Wang 2002)


  1. Song-chun Zhu, Cheng-en Guo, Yingnian Wu and  Yizhou Wang, "What are Textons?" (.pdf 767K), Proc. of 7th European Conference on Computer Vision, Copenhagen, Denmark, 2002.
  2. Song-Chun Zhu, Cheng-en Guo, Yizhou Wang, and Zijian Xu, "What are Textons?" (.pdf 1.7M), International Journal of Computer Vision. (accepted).