issues
search
Epiphqny
/
CondInst
Conditional Convolutions for Instance Segmentation, achives 37.1mAP on coco val
https://arxiv.org/abs/2003.05664
138
stars
15
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
However, the result of the original paper is 38.6map
#10
Liutingjin
opened
2 years ago
0
About the gpu memory issue
#9
NNNNAI
opened
3 years ago
2
FloatingPointError: Loss became infinite or NaN at iteration=772!
#8
xxxmian
opened
3 years ago
0
The file path was not found
#7
cty-xiaobai
opened
4 years ago
0
Only square mask predictions are supported
#6
xllau
opened
4 years ago
4
Where is the pre-trained weights?
#5
xllau
closed
4 years ago
2
Add __pycache__ to gitignore
#4
lyyyuna
closed
4 years ago
1
Some questions about the code.
#3
Yuxin-CV
closed
4 years ago
7
Postprocess of ContInst
#2
Anikily
closed
4 years ago
0
On the bilinear in your implementation.
#1
tianzhi0549
closed
4 years ago
29