mIOU=80.02 on cityscapes. My implementation of deeplabv3+ (also know as 'Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation' based on the dataset of cityscapes).
Hi, I noticed you updated your evluate.py. But I think your previous implementation is right, since if you use the evaluation code from https://github.com/mcordts/cityscapesScripts, you can get a mIou ~78.x for the single scale evaluation(same with your previous result). I use your code and model to do the evaluation on Cityscape test dataset. After submitting the test results to https://www.cityscapes-dataset.com/submit/ I got:
So I think maybe there is something wrong in your current evaluation code. But I cannot find out the reason so far.
LetsGoFir
commented
4 years ago
Hi, I noticed you updated your evluate.py. But I think your previous implementation is right, since if you use the evaluation code from https://github.com/mcordts/cityscapesScripts, you can get a mIou ~78.x for the single scale evaluation(same with your previous result). I use your code and model to do the evaluation on Cityscape test dataset. After submitting the test results to https://www.cityscapes-dataset.com/submit/ I got:
So I think maybe there is something wrong in your current evaluation code. But I cannot find out the reason so far.
Could I know if you have any clues?