Multiple Texture Boltzmann Machines
Jyri Kivinen and Christopher Williams
In: 15th International Conference on Artificial Intelligence and Statistics, 21-23 April 2012, La Palma, Canary Islands.
We assess the generative power of the mPoT-model of  with tiled-convolutional weight sharing as a model for visual textures by specically training on this task, evaluating model performance on texture synthesis and inpainting tasks using quantitative metrics. We also analyze the relative importance of the mean and covariance parts of the mPoT model by comparing its performance to those of its subcomponents,
tiled-convolutional versions of the PoT/FoE and Gaussian-Bernoulli restricted Boltzmann machine (GB-RBM). Our results suggest that while state-of-the-art or better performance can be achieved using the mPoT, similar performance can be achieved with the mean-only model. We then develop a model
for multiple textures based on the GB-RBM, using a shared set of weights but texture-specic hidden unit biases. We show comparable performance of the multiple texture model to individually trained texture models.