zhiheLu/CWT-for-FSS
Official code for "Simpler is Better: Few-shot Semantic Segmentation with Classifier Weight Transformer. ICCV2021".
PythonGPL-3.0
Issues
- 2
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:1 and cuda:0! (when checking argument for argument weight in method wrapper__cudnn_convolution)
#12 opened by lixiang007666 - 1
- 2
- 5
- 1
Can not achieve the results in the article
#19 opened by 7M7L - 11
- 1
loss nan ,miou 0
#24 opened by TungChintao - 0
Predict
#23 opened by yingqiluoCSDN - 0
replace my own dataset
#22 opened by yingqiluoCSDN - 2
Dataset replacement
#21 opened by yingqiluoCSDN - 0
Algorithm running speed
#20 opened by yingqiluoCSDN - 1
- 3
Script for Segmentation Map Visualization
#17 opened by kilickaya - 0
Evaluation on other dataset
#15 opened by viethoang303 - 1
Error run training model
#13 opened by viethoang303 - 1
Pretrained models link does not work
#11 opened by rasoulMrz - 1
pre-trained model
#9 opened by LIUYUANWEI98 - 2
training ppm by myself
#4 opened by ZS123-lang - 1
coco datasets
#5 opened by xiewende - 1
Can I ask for pre-train code?
#2 opened by mai556 - 1
Pascal SegmentationClassAug dataset
#7 opened by tt622 - 1
how to generate SegmentationClassAug?
#8 opened by WHL182 - 0
The pre-trained model
#3 opened by zwy1996 - 1
Checkpoint model for the CWT
#1 opened by yilinliu77