Probabilistic Diffusion for Interactive Image Segmentation

This paper presents an interactive image segmentation approach in which we formulate segmentation as a probabilistic estimation problem based on the prior user intention. Instead of directly measuring the relationship between pixels and labels, we first estimate the distances between pixel pairs and...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on image processing 2019-01, Vol.28 (1), p.330-342
Main Authors: Wang, Tao, Yang, Jian, Ji, Zexuan, Sun, Quansen
Format: Article
Language:English
Subjects:
Computing time
Diffusion
Feature extraction
Image segmentation
Interactive image segmentation
Iterative methods
Labels
likelihood learning
Manifolds
Optimization
paired distance measurement
Pixels
probabilistic estimation
Probabilistic logic
Probability theory
Robustness
Sun
unary potentials
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents an interactive image segmentation approach in which we formulate segmentation as a probabilistic estimation problem based on the prior user intention. Instead of directly measuring the relationship between pixels and labels, we first estimate the distances between pixel pairs and label pairs using a probabilistic framework. Then, binary probabilities with label pairs are naturally converted to unary probabilities with labels. The higher order relationship helps improve the robustness to user inputs. To improve segmentation accuracy, a likelihood learning framework is proposed to fuse the region and the boundary information of the image by imposing a smoothing constraint on the unary potentials. Furthermore, we establish an equivalence relationship between likelihood learning and likelihood diffusion and propose an iterative diffusion-based optimization strategy to maintain computational efficiency. Experiments on the Berkeley segmentation data set and Microsoft GrabCut database demonstrate that the proposed method can obtain better performance than the state-of-the-art methods.
ISSN:1057-7149
1941-0042
DOI:10.1109/TIP.2018.2867941