To achieve a highly realistic output conditioned on an exemplar patch, we propose a novel loss function that combines ideas from both style transfer and generative adversarial networks.

Deep neural networks (DNN) are commonly used to denoise and sharpen X-ray computed tomography (CT) images with the goal of reducing patient X-ray dosage while maintaining reconstruction quality. However, naive application of DNN-based methods can result in image texture that is undesirable in clinical applications. Alternatively, generative adversarial network (GAN) based methods can produce appropriate texture, but naive application of GANs can introduce inaccurate or even unreal image detail. In this paper, we propose a texture matching generative adversarial network (TMGAN) that enhances CT images while generating an image texture that can be matched to a target texture. We use parallel generators to separate anatomical features from the generated texture, which allows the GAN to be trained to match the desired texture without directly affecting the underlying CT image. We demonstrate that TMGAN generates enhanced image quality while also producing image texture that is desirable for clinical application.

Algorithmic We study the problem of example-based procedural texture information theory defines Kolmogorov complexity synthesis using highly compact models. Given a sample [17, 42] as the shortest computer program generating image, we use differentiable programming to train a a certain output. There exist parallels to the neighbouring generative process, parameterised by a recurrent Neural fields of compression and information theory, which Cellular Automata (NCA) rule. Contrary to the common handle the more specific instance of transmitting information belief that neural networks should be significantly overparameterised, using the fewest number of bits. Outside academia, we demonstrate that our model architecture work in the demo-scene [41] has established a long tradition and training procedure allows for representing complex texture of encoding complex geometries and rendering algorithms patterns using just a few hundred learned parameters, into extremely short programmatic code and small making their expressivity comparable to hand-engineered compiled binaries.