A biologically-inspired automatic matting method based on visual attention

Authors:
Wei Sun;Siwei Luo;Lina Wu
Affiliations:
School of Computer and Information Technology, Beijing Jiaotong University, Beijing, China;School of Computer and Information Technology, Beijing Jiaotong University, Beijing, China;School of Computer and Information Technology, Beijing Jiaotong University, Beijing, China
Venue:
ISNN'10 Proceedings of the 7th international conference on Advances in Neural Networks - Volume Part II
Year:
2010

Citing 6
Cited 0

Modeling visual attention via selective tuning

Artificial Intelligence - Special volume on computer vision
Attentional Selection for Object Recognition A Gentle Way

BMCV '02 Proceedings of the Second International Workshop on Biologically Motivated Computer Vision
An Iterative Optimization Approach for Unified Image Segmentation and Matting

ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
Selective Attention in the Learning of Invariant Representation of Objects

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Workshops - Volume 03
A Closed Form Solution to Natural Image Matting

CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1
2006 Special Issue: Modeling attention to salient proto-objects

Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Image matting is an important task in image and video editing In this paper we propose a novel automatic matting approach, which can provide a good set of constraints without human intervention We use the attention shift trace in a temporal sequence as the useful constraints for matting algorithm instead of user-specified “scribbles” Then we propose a modified visual selective attention mechanism which considered two Gestalt rules (proximity & similarity) for shifting the processing focus Experimental results on real-world data show that the constraints are useful Distinct from previous approaches, the algorithm presents the advantage of being biologically plausible.