DeepLayout icon indicating copy to clipboard operation
DeepLayout copied to clipboard
verified publisher icon kampta

Has anyone successfully trained on COCO?

Open sbkim052 opened this issue 4 years ago • 5 comments

Hello, thank you for sharing this code. I used the code with no changes and when I saw the visualization of the generated boxes are not really good. image image

Has anyone successfully trained on COCO?

sbkim052 avatar Dec 01 '21 06:12 sbkim052

Hello. Which coco annotations did you use?

  • 2017 Train/Val annotations [241MB]
  • 2017 Stuff Train/Val annotations [1.1GB]
  • 2017 Panoptic Train/Val annotations [821MB]

yuexihang avatar Dec 03 '21 09:12 yuexihang

Hello @yuexihang I used stuff_annotations_trainval2017.zip [1.1GB] Thanks for the reply:)

sbkim052 avatar Dec 03 '21 10:12 sbkim052

Hello @sbkim052 The annotation stuff_annotations_trainval2017.zip [1.1GB] has only 92 kinds of categoris. But the paper said they use the dataset with 171 annotations.

You can try to use the stuff and coco annotations together.

yuexihang avatar Dec 05 '21 09:12 yuexihang

Hi, according to the fifth page of this paper, it seems that the authors used the Panoptic version of COCO

JingyeChen avatar Dec 08 '22 14:12 JingyeChen

Hi, according to the fifth page of this paper, it seems that the authors used the Panoptic version of COCO

Hello,I use the panoptic version of coco,I modified part of the code in dataset.py to adapt to the corresponding version，the result is not right.

JackW987 avatar Feb 06 '23 02:02 JackW987