issues
search
yinghdb
/
EmbedMask
EmbedMask: Embedding Coupling for One-stage Instance Segmentation
MIT License
160
stars
19
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
New version
#18
JyunYuLai
closed
3 years ago
2
Positive samples for proposal embedding and margin
#17
JyunYuLai
closed
3 years ago
2
focal loss parameters
#16
gewenbin292
closed
4 years ago
1
Why margin need to be scaled by "scale_size"?
#15
shoutOutYangJie
opened
4 years ago
9
Question about lovasz_hinge_loss
#14
Epiphqny
closed
4 years ago
2
Embedding Visualization
#13
eugenelawrence
closed
4 years ago
1
Speed performance
#12
Epiphqny
closed
4 years ago
2
About comparison with state-of-the-art methods
#11
mychina75
opened
4 years ago
3
Bugs in get_pos_proposal_indexes for mask loss
#10
Devin147
closed
4 years ago
1
reformulate embedmask based on detectron2
#9
TengFeiHan0
closed
4 years ago
11
about smooth_loss in training
#8
UdonDa
closed
4 years ago
2
unreachable gradient
#7
afantideng
closed
4 years ago
5
Smooth loss is 0.0000 when Training with one GPU
#6
SkeletonOne
closed
4 years ago
2
Performance
#5
trungpham2606
opened
4 years ago
7
Nan for classification loss in training stage.
#4
wangyuxin87
closed
4 years ago
1
FCOS structure
#3
trungpham2606
closed
4 years ago
3
Why so many outputs
#2
lucasjinreal
closed
4 years ago
0
Train Speed
#1
RainHxj
closed
4 years ago
1