Viet-Anh on Software Logo

What is: TSRUs?

SourceTransformation-based Adversarial Video Prediction on Large-Scale Data
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

TSRUs, or Transformation-based Spatial Recurrent Unit p, is a modification of a ConvGRU used in the TriVD-GAN architecture for video generation.

It largely follows TSRUc, but computes each intermediate output in a fully sequential manner: like in TSRUc, cc is given access to h^_t1\hat{h}\_{t-1}, but additionally, uu is given access to both outputs h^_t1\hat{h}\_{t-1} and cc, so as to make an informed decision prior to mixing. This yields the following replacement for uu:

u=σ(W_u_n[h^_t1;c]+b_u)u = \sigma\left(W\_{u} \star\_{n}\left[\hat{h}\_{t-1};c\right] + b\_{u} \right)

In these equations σ\sigma and ρ\rho are the elementwise sigmoid and ReLU functions respectively and the _n\star\_{n} represents a convolution with a kernel of size n×nn \times n. Brackets are used to represent a feature concatenation.