File Exchange

image thumbnail

Semantic segmentation using Pascal VOC

version 1.0.1 (2.23 MB) by Kenta
This example shows how to train a semantic segmentation network using deep learning using Pascal VOC dataset.

24 Downloads

Updated 24 May 2020

View License

[English]
This example shows how to train a semantic segmentation network using deep learning. This example was a modified version of the Matlab official document entitled Semantic Segmentation Using Deep Learning [1]. A semantic segmentation network classifies every pixel in an image, resulting in an image that is segmented by class as shown in the thumbnail. To illustrate the training procedure, this example trains Deeplab v3+ [2], one type of convolutional neural network (CNN) designed for semantic image segmentation. This example uses the dataset in Visual Object Classes Challenge 2012 (VOC2012) [3]. To run this code, please down load the dataset available at [3].

[Japanese]
この例は、matlabの公式ドキュメント[1]をもとにセマンティックセグメンテーションを行う例を示します。
セグメンテーションでは、画像の各画素(ピクセル)ごとに、そのピクセルはどのクラスか(猫か、車か、空か、など)の分類を行います。
それによって、サムネイルにあるような詳細な推論を行うことができます。サムネイルはGIFファイルになっていますので、クリックすると動画のように再生されます。deep lab v3とよばれるネットワークを利用し、pascal VOC [3] というデータセットのセグメンテーションを行います。
[Key words]
deep learning, pascal VOC, pixel-wise, segmentation, semantic segmentation

Reference
[1] Semantic Segmentation Using Deep Learning
(https://jp.mathworks.com/help/vision/examples/semantic-segmentation-using-deep-learning.html?lang=en)
[2] Chen, Liang-Chieh et al. “Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation.” ECCV (2018).
[3] Visual Object Classes Challenge 2012 (VOC2012)
(http://host.robots.ox.ac.uk/pascal/VOC/voc2012/index.html)

Cite As

Kenta (2020). Semantic segmentation using Pascal VOC (https://www.mathworks.com/matlabcentral/fileexchange/75938-semantic-segmentation-using-pascal-voc), MATLAB Central File Exchange. Retrieved .

Comments and Ratings (2)

Updates

1.0.1

Japanese description added

MATLAB Release Compatibility
Created with R2020a
Compatible with any release
Platform Compatibility
Windows macOS Linux