Error in Semantic Segmentation Using Deep Learning Demo

Hello everyone,

I'm following the example here: https://www.mathworks.com/help/vision/ug/semantic-segmentation-using-deep-learning.html

I ran into an error at this line:

pximds = pixelLabelImageDatastore(imdsTrain,pxdsTrain, ...
     'DataAugmentation',augmenter);

The error message is this:

Error using images.internal.datastore.PixelLabelImageDatastore (line 244)

pixelLabelImageDatastore supports ColorPreprocessing, DataAugmentation and OutputSize properties only for 2-D

images. See the transform method to resize, augment or modify 3D data.

Error in pixelLabelImageDatastore (line 195)

this = [e-mail](varargin{:});

I've checked the images in my CamVid folder and all of them are 2-D PNG files, as downloaded. I have not changed the augmenter line before in anyway. May I know what is causing the error? Thank you!

[Madhav Thakker]

Hi JW,

augmenter = imageDataAugmenter('RandXReflection',true,...

'RandXTranslation',[-10 10],'RandYTranslation',[-10 10]);

Did you change the definition?

[Melvin Mathew]

I have the same issue. Was this resolved in any way?