File Exchange

image thumbnail

3-D Deep Learning : 3-D Volume Labeling Assist Tool

version 1.0.0 (325 KB) by Kei Otsuka
Interactively Create Mask, and Semi-Automatically Label Volume Data Using 3-D Active Contour on the Labeler App


Updated 03 Jul 2019

View License

Semantic segmentation (3-D Deep Learning) is really powerful approach to segment specific 3-D regions in an volume data. Good ground truth data is crucial for training deep neural network, but creating and maintaining a diverse and high-quality set of annotated data requires significant effort.
The videoLabeler and imageLabeler app makes this process easy and efficient, but these apps are designed for 2-D images or video. For those who want to extend the capability of Labeler apps to 3-D volume data, this tool give an idea about how to add custom algorithms for 3-D volume data to the app.
As an example, this tool enable you to label XZ slice images on another figure window and then you can use active contour method on this tool to segment target object by using labels as an initial mask.
Segmented object is converted to 3-D labeled volume data and back to Labeler App.

videoLabelerやimageLabelerを利用することで、このラベリングの作業を効率的に行うことができますが、これらのツールは主に2-D画像データや動画を扱うことを前提としています。これらのLabeler Appを3-Dボリュームデータにも活用したい方のために、カスタムの半自動化アルゴリズムを3-Dボリュームデータ用に準備してAppに追加するための例をご紹介します。
3-Dボリュームデータに対してラベルを付与する場合、App上ではXY平面に対するラベリングとなりますが、このツールを利用することでXZ平面に対してもラベリングを可能とし、さらにActive Contourを利用して対象オブジェクトに対するラベリング作業を半自動化できます。


Cite As

Kei Otsuka (2021). 3-D Deep Learning : 3-D Volume Labeling Assist Tool (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (2)

Shahid Khan

I want to use two separate labels i.e. Left Lung and Right Lung. But it gives an error
"Error using labelvolshow
Expected LabelColor to be of size 3x3, but it is of size 2x3"

Any idea how to fix it?

Ryuichi Nakahara

MATLAB Release Compatibility
Created with R2019a
Compatible with R2019a and later releases
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!