Problem Converting Categorical Array to Uint8

I have a dataset of 8bit label images that Im trying to resize to achieve consistency in the dataset, I tried using the supporting function resizeCamVidPixelLabels provided in the Semantic Segmentation example https://uk.mathworks.com/help/vision/examples/semantic-segmentation-using-deep-learning.html#d119e463 , however the output I am getting does not seem to preserve the class label data. The output images I am getting are as shown in the Image below

whereas the original images are like this.

Any idea what might seem to be the problem?

P.S: both images shown below arnt the same, they are different images but you get the idea