Random Field Model for Integration of Local Information and Global Information

Authors:
Takahiro Toyoda;Osamu Hasegawa
Affiliations:
Tokyo Institute of Technology, Kanagawa-ken;Tokyo Institute of Technology, Yokohama
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
2008

Citing 0
Cited 6

The relative potential field as a novel physics-inspired method for image analysis

WSEAS Transactions on Computers
Graph cut based inference with co-occurrence statistics

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part V
A novel virtual potential field transform for gray-scale images inspired by physical electro-static field

ICCOMP'10 Proceedings of the 14th WSEAS international conference on Computers: part of the 14th WSEAS CSCC multiconference - Volume II
Estimation of 3-D human body posture via co-registration of 3-D human model and sequential stereo information

Applied Intelligence
Local label descriptor for example based semantic image labeling

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part VII
Inference Methods for CRFs with Co-occurrence Statistics

International Journal of Computer Vision

Quantified Score

Hi-index	0.14

Visualization

Abstract

This paper presents a proposal of a general framework that explicitly models local information and global information in a conditional random field. The proposed method extracts global image features as well as local ones and uses them to predict the scene of the input image. Scene-based top-down information is generated based on the predicted scene. It represents a global spatial configuration of labels and category compatibility over an image. Incorporation of the global information helps to resolve local ambiguities and achieves locally and globally consistent image recognition. In spite of the model's simplicity, the proposed method demonstrates good performance in image labeling of two datasets.