search

XiaLiPKU / EMANet

The code for Expectation-Maximization Attention Networks for Semantic Segmentation (ICCV'2019 Oral)
https://xialipku.github.io/publication/expectation-maximization-attention-networks-for-semantic-segmentation/
GNU General Public License v3.0
680 stars 130 forks source link

How can eval.py generate results of segmentation? #13

Closed illutheplanet closed 4 years ago

illutheplanet commented 5 years ago

When running eval.py only appears mIoU, what should I do to get the segmentation maps?

XiaLiPKU commented 5 years ago

When running eval.py only appears mIoU, what should I do to get the segmentation maps?

I would like to add this function further. For now, you can hack the codes, and save the 'pred' into a png. Moreover, to visualize the results by eye, you may need colorize the generated pngs.

robotcv commented 4 years ago

how to show attention map?

XiaLiPKU commented 4 years ago

how to show attention map?

Just output the Z is enough. But for better color and value, you would better normalize the value map.

robotcv commented 4 years ago

how to show attention map?

Just output the Z is enough. But for better color and value, you would better normalize the value map.

robotcv commented 4 years ago

how to show attention map?

Just output the Z is enough. But for better color and value, you would better normalize the value map.

Okay, thank you very much