-
I have studied the tutorials and Readme of both AdelaiDet and Detectron2. However, I am still not quite clear how to write a trainer for SoloV2. I am quite new to Detectron2, but I know Pytorch well. …
-
微博内容精选
-
希望能够支持Boxinst，我测试了效果很好，可以解决工业标注的问题。麻烦考虑一下呀！！！！！
https://github.com/aim-uofa/AdelaiDet/blob/master/configs/BoxInst/README.md
BoxInst: High-Performance Instance Segmentation with Box Annotations
-
Greetings! Thanks for supporting your code for your awesome work!
I noticed that you have listed the result of BoxInst in your paper, would you have try to re-implement BoxInst in your mmdet code?
-
I followed QuickStart in BlendMask's README.
```bask
wget -O blendmask_r101_dcni3_5x.pth https://cloudstor.aarnet.edu.au/plus/s/vbnKnQtaGlw8TKv/download
python demo/demo.py \
--config-file con…
-
Hi! Thanks for opensourcing the code.
Both work use the segmentation loss for extra supervision to the mask feature(blender).
What is difference of these two?
From the code, I found that Blen…
lxtGH updated
2 years ago
-
Thank you for your great job! I would like to ask you about the redesigned mask loss in BoxInst. Its pairwise term is calculated according to formula (8) or formula (5)?
-
Thanks for making this great work open source.
but I have trained 255K steps from scratch , and the totaltext hmean is always 0. (Use 8 GPUS,Pretrain/testr_r50.yaml and all the parameters are def…
-
In DynamicMaskHead > __call__
```
if len(pred_instances) == 0:
loss_mask = mask_feats.sum() * 0 + pred_instances.mask_head_params.sum() * 0
loss_mask["loss_mask"] = loss_mask
if self.boxins…