PixelCNN Models with Auxiliary Variables for Natural Image Modeling
This addresses image generation quality and efficiency for computer vision applications, though it appears incremental as an extension of existing PixelCNN architectures.
The authors tackled PixelCNN models' limitations in capturing high-level image information and slow sampling by incorporating auxiliary variables like grayscale views or image pyramids, resulting in more realistic image samples than previous state-of-the-art models.
We study probabilistic models of natural images and extend the autoregressive family of PixelCNN architectures by incorporating auxiliary variables. Subsequently, we describe two new generative image models that exploit different image transformations as auxiliary variables: a quantized grayscale view of the image or a multi-resolution image pyramid. The proposed models tackle two known shortcomings of existing PixelCNN models: 1) their tendency to focus on low-level image details, while largely ignoring high-level image information, such as object shapes, and 2) their computationally costly procedure for image sampling. We experimentally demonstrate benefits of the proposed models, in particular showing that they produce much more realistically looking image samples than previous state-of-the-art probabilistic models.