-
ghost updated
5 years ago
-
How do I modify the object detection demo app to use my model trained with faster-rcnn-r50. Currently the app only supports picodet and ssd models right?
-
I trained the network from scratch with a dataset. I want to perform finetuning with different number of classes. Do you have any suggestions? @jwyang
-
Hi,
in the rbg's py-faster rcnn, he un-normalize the bounding box regression before saving the weight (https://github.com/rbgirshick/py-faster-rcnn/blob/master/lib/fast_rcnn/train.py#L70).
But I…
twmht updated
5 years ago
-
# 重要模块
## Region Proposal network
- [ ] /home/ncrc-super/anaconda3/envs/py17/lib/python3.7/site-packages/torchvision/models/detection/rpn.py
- [ ] /home/ncrc-super/anaconda3/envs/py17/lib/pytho…
-
When I run the steps as _Preparing for Testing_ ,the faster_rcnn_build.m file compiles failed ,the matlab(2015b) showed :-1:
"Error using mex
MEX cannot find library 'cudart' specified with the -l op…
-
Hi.
I'm glad Microsoft team integrated Faster-RCNN to windows caffe.
But I can not see any samples.
Would you provide any samples? Especially, I want to check prototxt to use Faster-RCNN.
-
Can someone tell me why we have to include "bg" class? I am working on an medical image dataset. I am not sure whether I have to include some "bg" class in my training dataset.
This is what I though…
-
在尝试导出onnx格式,出现不支持的OPS:
https://github.com/PaddlePaddle/PaddleX/blob/develop/docs/appendix/model_zoo.md
目标检测模型
![image](https://user-images.githubusercontent.com/99886928/169239432-adf2b2f6-4220-43c…
-
### Description
记录开发调试过程和遇到的问题。