long123524 / seanet_torch Goto Github PK
View Code? Open in Web Editor NEWISPRS: Using a semantic edge-aware multi-task neural network to delineate agricultural parcels from remote sensing images
ISPRS: Using a semantic edge-aware multi-task neural network to delineate agricultural parcels from remote sensing images
Dear authors,
I trivially copied the file and instantiated the model as:
torch_device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
print('Using device:', torch_device)
model = SEANet(num_classes=1) # the same error with a different num_classes
print(summary(model, (8, 3, 512, 512))) # the same error with a different batch_size
I got this error:
343 def forward(self, x):
344 # VGG
345 img_H, img_W = x.shape[2], x.shape[3]
-> 347 conv1_1 = self.activ(self.conv1_1(x))
RuntimeError: Expected 3D (unbatched) or 4D (batched) input to conv2d, but got input of size: [2, 8, 3, 512, 512]
Could you help me?
Many thanks for your excellent work.
D
Hello,
I can not access the pretrained weights through the current link.
can you provide an alternative link for international people, please.
您好,有个问题一直困惑我,我看到您论文的结果是有地块边界的,但是我试验了把mask的图像换成边界(把mask换成contour),并没法预测。我也试过直接用边界线当mask,但是这样的mask无法计算dist_contour和contour,也无法进行训练。如果我用面去预测,就只能得到一整块地块的边界,所以我想请问您,如何能得到一小块一小块地块的边界?谢谢您!
如果我不用dist_contour和contour会有影响吗?因为我换了数据,生成dist_contour是有问题的,我在arcgispro上打开是灰色的整张图片,所以想去掉dist_contour和contour来试验
有两个问题想请教一下
1、预测的第二个输出是[6,1,256,256] 怎么获取边缘的输出
2、预测的mask还是存在地块的粘连,怎么通过后处理获取最终的地块矢量呢 谢谢
请问一下,您有没有可以将预测结果又拼回一整张图片的代码,也就是cai_duobands.py的逆过程,谢谢!
您好!我想将mask的数据换成contour的数据,以实现边界的提取,但是我现在预测输出的结果是错误的(如何用原mask,是可以提取出地块的),请问这个模型能否实现边界提取的操作呢?谢谢您!!!
论文中的数据集可以分享下吗,仅用于研究使用
@long123524
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.