CamVid-UNet accuracy measurement

[lx2u16]

Hi,

I can see there is an UNet demo that is trained and tested using the CamVid dataset. I think it is doing semantic segmentation which the accuracy is typically measured by IoU.

However, your script shows a top1 accuracy. I wonder what it is?

Thanks.

[ermanok]

The dataset gives a segmentation map, where each pixel is assigned to one of the target classes, of the image as the label. So we use pixel-wise classification accuracy as the Top-1 accuracy.

[lx2u16]

@ermanok thanks, and the final Top-1 accuracy is the mean pixel-wise classification accuracy of the entire test dataset?

[ermanok]

that is correct.