issues
search
Lliu666
/
MSHNet
Implement of CVPR2024 'Infrared Small Target Detection with Scale and Location Sensitivity'
41
stars
10
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
为什么NUDT-SIRST这个数据集iou只有0.8，NDAnet原文不就有0.87吗，你论文中DNAnet只有0.79，请问你是怎么划分数据集的
#16
6017203257
opened
1 month ago
4
The left of Fig. 3 is wrong in the paper
#15
CongLi-18
opened
3 months ago
0
代码缺少
#14
Songsong0824
opened
4 months ago
0
提供的.tar和.pkl文件不能test
#13
gottastudy
opened
4 months ago
2
训练时，保存在metric.log中的PD值显示有问题（PD值是不断变小的），但是训练得到的权重测试得到的PD值是正确的
#12
LYC-1021
opened
5 months ago
5
在NUDT上训练，loss不是很稳定，代码看起来只有IRSTD-1k的数据集load，提供一下NUDT的代码以及数据集结构？
#11
Deaddawn
opened
5 months ago
5
用于检测的代码可否提供一下？
#10
yangyahu-1994
opened
5 months ago
0
您好，数据集结构是怎么样的呀？
#9
yangyahu-1994
closed
5 months ago
1
实验环境可否提供一下？
#8
yangyahu-1994
closed
5 months ago
1
新结果的参数可否提供一下，我设定的bs=4，lr=0.05
#7
JiangNan-Yang
opened
6 months ago
13
关于论文中decoder后的4个3*3的卷积和代码不对应，代码是1*1，请问是画错了还是代码有误
#6
GanGeeee
opened
6 months ago
1
Current Location Sensitive Loss form based on `torch.where` is not differentiable.
#5
lartpang
closed
6 months ago
1
模型训练时显示self.down()未定义
#4
wwwhzahu
closed
6 months ago
2
I want to know if the code will open?Thank you!
#3
icarus-llz
closed
6 months ago
2
Thanks when the code release
#2
MrDec
closed
6 months ago
0
Source code
#1
yeah-creater
closed
6 months ago
0