ISTR icon indicating copy to clipboard operation
ISTR copied to clipboard

ISTR: End-to-End Instance Segmentation with Transformers (https://arxiv.org/abs/2105.00637)

Results 23 ISTR issues
Sort by recently updated
recently updated
newest added

how can i train istr with custom dataset ?

While running the code: `python projects/ISTR/train_net.py --num-gpus 4 --config-file projects/ISTR/configs/ISTR-R50-3x.yaml --eval-only MODEL.WEIGHTS ./output/model_final.pth` I encountered the following error-No such directory found projects/ISTR/configs/ISTR-R50-3x.yaml . I looked into the configs folder and...

Great paper and really nice code! Would you mind adding a license file to the repository?

![image](https://user-images.githubusercontent.com/83260396/144528497-3bf96d2d-6fad-4671-a024-2c9ef91cbc00.png) 看作者文中使用ResNet50-FPN~在coco数据集的精度是38.6 ,但是我使用read me提供的训练口令 在ResNet50-FPN训练出来的结果是 35.9631,请问我是哪里操作有问题吗

AssertionError: /ISTR/projects/ISTR/istr/util/box_ops.py in generalized_box_iou(boxes1, boxes2) 49 # degenerate boxes gives inf / nan results 50 # so do an early check ---> 51 assert (boxes1[:, 2:] >= boxes1[:, :2]).all() 52...

请问作者4块gpu训练了几天

Hi, thanks for your awesome work. Have you tried the Mask Embeddings in DETR? I am very interested about the performance of this method on DETR. Thank you.

I didn't understand how to position embedding got in this paper. I read this article in detail, but I didn't find the answer. Could you please introduce it? Thank you.

SOLOv2 becomes slowest model in your papers, is that really? But as SOLOv2 official paper claims it should faster than most instance segmentation methods, at least compare with MaskRCNN.