-
**Mask R-CNN 论文详解**
https://zhuanlan.zhihu.com/p/57759536
=Faster-RCNN 的基础上添加一个分支网络(就是把ROI pooling替换为ROI align),在实现目标检测的同时,把目标像素分割出来。
![image](https://user-images.githubusercontent.com/47746159/821…
-
Great work!
I have a question about your CENet mentioned in paper.
Why did you use method of summation and 5 fc layers to handle five residual features rather than method of concat and 1 fc layer?
…
-
Hi, nice work! I have a question after reading your paper. Though you claim that Coordattention works well for mobile network, it seems long-range dependencies may also benifit the regular deep netwo…
-
LeNet-5, AlexNet, GoogleLeNet, VGGNet, ResNet-34, ResNet-50, Xception, SENet
-
New Treasure Vault mechanic has *so much potential*.
For Headhunters, they could actually collect heads! Promethean's could collect ancient philosophical treatises and essays. Hellenes could collec…
-
Hello,
Can you help me solve this issue. I am testing the pretrained weights and this is the error i am getting
![image](https://user-images.githubusercontent.com/56354373/124066435-45eaf200-da5…
-
hi, I think squeeze_excitation_layer should be added after inception block ,not inception_resnet block
Am I right ?
-
Hi, while reading your code, I found unnecessary line in m2det.py.
(Commint ID: f0decfe619347a451d282b10ac8b01fa3242adf9)
In m2det.py line 52.
```python
attention = tf.reduce_mean(feature, axis=…
-
This is a master issue to track requests for adding new pre-trained models to torchvision.
Here is the (potentially incomplete) list I compiled:
- [x] ResNext https://github.com/pytorch/vision/iss…
-
When I added the SE module of senet in yolov3. I have an illegal memory access:
Avg (IOU: 0.389989, GIOU: 0.206918), Class: 0.228798, Obj: 0.380333, No Obj: 0.492784, .5R: 0.500000, .75R: 0.000000,…