Question About Uncertainty Map Visualization in Your Paper

[UntrainedButLoveCode]

Hello, I really appreciate your excellent paper and your great work.

I have a question regarding the visualization of the two uncertainty maps in your paper. Could you please let me know what method you used to visualize them? If possible, would you be able to share the relevant code?

Below is the way I converted them into grayscale images for visualization:

aleatoric = outputs['aleatoric'][0].detach().cpu().numpy().astype(np.float32)
epistemic = outputs['epistemic'][0].detach().cpu().numpy().astype(np.float32)

cv2.imwrite(save_alea, _to_gray_uint8(aleatoric))
cv2.imwrite(save_epis, _to_gray_uint8(epistemic))

However, the results look very different from those shown in your paper. Could you please explain why there is such a large difference?

Thank you very much for your time and help.