Pretrained model on cityscapes

[codermckee]

Do you have any pretrained model on cityscapes or mapillary?

[codermckee]

I use the models pretrained on ImageNet or voc to train mapillary dataset, but get low mIoU. so sad.

[zhanghang1989]

Will add cityscapes in GluonCV in later release.

[codermckee]

Will add cityscapes in GluonCV in later release.

thanks for your reply. I am a beginner of semantic segmentation, and I have a problem which confused me. I use pretrained model on ImageNet to train object detector and I can get good mAP. But it doesn't work when it comes to semantic segmentation. would you mind giving me some advice? thank you very much!

[zhanghang1989]

There are many details in the implementation for reproducing SoTA result. Synchronized Batch Normalization is one of the most important key in semantic segmentation.

[zhanghang1989]

You may start with my repo https://hangzhang.org/PyTorch-Encoding/experiments/segmentation.html

[codermckee]

You may start with my repo https://hangzhang.org/PyTorch-Encoding/experiments/segmentation.html

Thank you very much !

[XiaLiPKU]

@zhanghang1989 Can you also add the pretrained model on COCO? I see that has been published in GluonCV.