TGHop: An Explainable, Efficient and Lightweight Method for Texture Generation
arXiv:2107.04020
Abstract
An explainable, efficient and lightweight method for texture generation, called TGHop (an acronym of Texture Generation PixelHop), is proposed in this work. Although synthesis of visually pleasant texture can be achieved by deep neural networks, the associated models are large in size, difficult to explain in theory, and computationally expensive in training. In contrast, TGHop is small in its model size, mathematically transparent, efficient in training and inference, and able to generate high quality texture. Given an exemplary texture, TGHop first crops many sample patches out of it to form a collection of sample patches called the source. Then, it analyzes pixel statistics of samples from the source and obtains a sequence of fine-to-coarse subspaces for these patches by using the PixelHop++ framework. To generate texture patches with TGHop, we begin with the coarsest subspace, which is called the core, and attempt to generate samples in each subspace by following the distribution of real samples. Finally, texture patches are stitched to form texture images of a large size. It is demonstrated by experimental results that TGHop can generate texture images of superior quality with a small model size and at a fast speed.
arXiv admin note: substantial text overlap with arXiv:2009.01376
References in corpus (4)
- Stable and Controllable Neural Texture Synthesis and Style Transfer Using Histogram Losses
- R-PointHop: A Green, Accurate, and Unsupervised Point Cloud Registration Method
- Successive Subspace Learning: An Overview
- VoxelHop: Successive Subspace Learning for ALS Disease Classification Using Structural MRI