Non-Gibbsian Markov Random Field Models for Contextual Labelling of Structured Scenes
نویسندگان
چکیده
In this paper we propose a non-Gibbsian Markov random field to model the spatial and topological relationships between objects in structured scenes. The field is formulated in terms of conditional probabilities learned from a set of training images. A locally consistent labelling of new scenes is achieved by relaxing the Markov random field directly using these conditional probabilities. We evaluate our model on a varied collection of several hundred handsegmented images of buildings.
منابع مشابه
Markov Random Fields with Asymmetric Interactions for Modelling Spatial Context in Structured Scene Labelling
In this paper we propose a Markov random field with asymmetric Markov parameters to model the spatial and topological relationships between objects in structured scenes. The field is formulated in terms of conditional probabilities learnt from a set of training images. A locally consistent labelling of new scenes is achieved by relaxing the Markov random field directly using these conditional p...
متن کاملCluster-Based Image Segmentation Using Fuzzy Markov Random Field
Image segmentation is an important task in image processing and computer vision which attract many researchers attention. There are a couple of information sets pixels in an image: statistical and structural information which refer to the feature value of pixel data and local correlation of pixel data, respectively. Markov random field (MRF) is a tool for modeling statistical and structural inf...
متن کاملContextual Classification of Image Patches with Latent Aspect Models
We present a novel approach for contextual classification of image patches in complex visual scenes, based on the use of histograms of quantized features and probabilistic aspect models. Our approach uses context in two ways: (1) by using the fact that specific learned aspects correlate with the semantic classes, which resolves some cases of visual polysemy often present in patch-based represen...
متن کاملEfficient 3D Scene Labelling Using Fields of Trees
Great progress on the major computer vision tasks has been reported over the last few years. SLAM systems such as PTAM and KinectFusion [5] create 3D models of unknown and uncontrolled environments in real time and structured and independent classification methods [6, 3] are pushing the frontiers in object recognition. However, establishing links between 3D reconstruction and object recognition...
متن کاملContextual Reclassification of Multispectral Images: A Markov Random Field Approach
This work presents methods for multispectral image classification using the contextual classifiers based on Markov Random Field (MRF) models. Performance of some conventional classification methods is evaluated, through a Monte Carlo study, with or without using the contextual reclassification. Spatial autocorrelation is present in the computer-generated data on a true scene. The total misclass...
متن کامل