Synthesising Textures Using Variable Neighbourhood Searching
نویسندگان
چکیده
Texture synthesis aims to define and reproduce discriminating image features. These features are used to associate with and differentiate between two textures. Often texture is composed of a pattern with an element of randomness in each feature’s appearance, position and orientation. The goal is to imitate the sample texture in such a way that sample and synthesised texture are perceived to be generated by the same source. One of the recent texture synthesis methods selects an output pixel by searching with its already generated neighbourhood for a corresponding match in the sample image. The neighbourhood size is fixed and blurring of texture features often results. Our method attempts to avoid this problem by enabling a dynamic, accelerated neighbourhood search. The window size varies with each output pixel and is determined by the current neighbourhood intensity configuration.
منابع مشابه
Texture Synthesis via a Non-parametric Markov Random Field
In this paper we present a non-causal non-parametric multiscale Markov random field (MRF) texture model that is capable of synthesising a wide variety of textures. The textures that this model is capable of synthesising vary from the highly structured to the stochastic type and include those found in the Brodatz album of textures. The texture model uses Parzen estimation to estimate the conditi...
متن کاملA Nonparametric Multiscale Markov Random Field Model for Synthesising Natural Textures
In this paper we present a non-causal, non-parametric, multiscale, Markov random field (MRF) texture model. The model is capable of capturing the characteristics of and synthesising a wide variety of textures, varying from the highly structured to the stochastic. We introduce a novel multiscale texture synthesis algorithm that allows us to use large neighbourhood systems to model some complex n...
متن کاملTexture synthesis and unsupervised recognition with a nonparametric multiscale Markov random field model
In this paper we present noncausal, nonparametric, multiscale, Markov Random Field (MRF) model for synthesising and recognising texture. The model has the ability to capture the characteristics of a wide variety of textures, varying from the structured to the stochastic. For texture synthesis, we use our own novel multiscale approach, incorporating local annealing, allowing us to use large neig...
متن کاملTexture Synthesis and Unsupervised Recognition with Nonparametric Multiscale Markov Random Field Models
In this paper we present noncausal, nonparametric, multiscale, Markov Random Field (MRF) models for synthesising and recognising texture. The models have the ability to capture the characteristics of a wide variety of textures, varying from the structured to the stochastic. For texture synthesis, we use our own novel multiscale approach, incorporating local annealing, allowing us to use large n...
متن کاملSynthesising Dynamic Textures using Convolutional Neural Networks
Here we present a parametric model for dynamic textures. The model is based on spatiotemporal summary statistics computed from the feature representations of a Convolutional Neural Network (CNN) trained on object recognition. We demonstrate how the model can be used to synthesise new samples of dynamic textures and to predict motion in simple movies.
متن کامل