CSRNet-pytorch

CSRNet-pytorch copied to clipboard

请问一下，在csrnet那篇论文中，最后有psnr和ssim的比较，源网络不是有3次池化吗，那么最后怎么能比较psnr和ssim呢?

eain3314

worldexpo'10 roi problem

3

jiezishu737

求问如何选择的val数据集,有什么特殊的标准吗

3

Amelie01

High inference time in comparison to keras implementation

1

As stated by the issue title. The [CSRNet-keras](https://github.com/Neerajj9/CSRNet-keras) implementation is a lot faster than this implementation. In my case, one single inference on a Full HD image required ~6100 ms....

Pipoderoso

IOError: Unable to create file

2

Hi, @leeyeehoo When I running two pieces of training, I encountered this issue. Traceback (most recent call last): File "train.py", line 249, in main() File "train.py", line 99, in main...

xyyu-callen

首先。图像预处理是指将所有待训练和测试的图片根据标注的人头点的坐标生成对应的密度图，并与标注的人头总数一起作为ground-truth。训练阶段是指将所有的训练集（包括图像预处理生成的ground-truth）输送到以VGG16前十层作为前端的网络进行人头特征提取，然后将提取到的人头特征输送到空洞卷积神经网络的进行训练，最后通过提取出的人头位置特征生成对应的密度图。 麻烦看一下我理解的对吗？谢谢

eain3314

Perform data augmentation in training and normalize correctly at test…

4

Working with the code I noticed that data augmentation (i.e. random crop and random flip) was not performed, and that in `val.ipynb` data normalisation at test time was not working...

covix

kesharwaniarya

Part_B上训练模型不收敛

12

环境：win10+ cuda9.0 +Pytorch4.0 （GTX1070） 加载了VGG16的预训练参数 直接加8倍upsample模型不收敛 不加upsample，MAE一直在68左右，lr改过1e-6，1e-7，也不知道怎么回事。 有没有什么训练日志什么的，或者训练上的trick

libaikai

Can we use our photos and print density map?

2

Hello, I'm a researcher with my team at NCTU, Taipei, Taiwan. I face a problem. Whenever I want to replace the photos in the dataset with other pictures, I run...

tang0809