ricozero / binary-segmentation-evaluation-tool

This repo is developed for evaluating binary image segmentation results. Measures, such as MAE, Precision, Recall, F-measure, PR curves and F-measure curves are included.

Binary-Segmentation-Evaluation-Tool

This repo is developed for the evaluation of binary image segmentation results.

The Code was used for evaluation in CVPR 2019 paper 'BASNet: Boundary-Aware Salient Object Detection code', Xuebin Qin, Zichen Zhang, Chenyang Huang, Chao Gao, Masood Dehghan and Martin Jagersand.

Contact: xuebin[at]ualberta[dot]ca

Required libraries

Python 3.6.6 (version newer than 3.0)

numpy 1.15.2

scikit-image 0.14.0

matplotlib 2.2.3

Usage

Please follow the scripts in quan_eval_demo.py

Implemented measures

1. MAE Mean Absolute Error

2. Precision, Recall, F-measure (This is the python implementation of algorithm in sal_eval_toolbox)

3. Precision-recall curves

4. F-measure curves

Future measures

IoU Intersection-over-Union

relax boundary F-measure

...

Citation

@InProceedings{Qin_2019_CVPR,
author = {Qin, Xuebin and Zhang, Zichen and Huang, Chenyang and Gao, Chao and Dehghan, Masood and Jagersand, Martin},
title = {BASNet: Boundary-Aware Salient Object Detection},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}