haotian/duan8
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Improve DBNet README.md (#467)

Browse Source 
* Fix: syntax error in common.hpp , cuda & tensorrt not included in CMakeLists.txt

* Fix: layer warining while building in common.hpp and denet.cpp

* Fix: NOT droping mini boxes and low score boxes.

* Fix: The prediction boxes are NOT expanded by the specified proportion, which will lead to serious errors

* Improve DBNet README

* Improve DBNet README

* Improve DBNet README
This commit is contained in:
PeterH0323 2021-04-08 10:52:33 +08:00 committed by GitHub
parent 66d641814e
commit 12f7c1b8b1
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 19 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
dbnet/README.md
View File

@ -3,19 +3,20 @@
The Pytorch implementation is [DBNet](https://github.com/BaofengZan/DBNet.pytorch).
<p align="center">
<img src="https://user-images.githubusercontent.com/20653176/100968101-b044be00-356b-11eb-808c-9597cbe1f8de.jpg">
<img src="https://user-images.githubusercontent.com/25873202/113959270-1eb8c600-9855-11eb-9c4d-1e6dc8e38a17.jpg">
</p>
## How to Run
* 1. generate .wts
* 1. generate `.wts`
Download code and model from [DBNet](https://github.com/BaofengZan/DBNet.pytorch) and config your environments.
In tools/predict.py, set `save_wts` as `True`, and run, the .wts will be generated.
Go to file`tools/predict.py`, set `--save_wts` as `True`, then run, the `DBNet.wts` will be generated.
onnx can also be exported, just need to set `onnx` as `True`.
Onnx can also be exported, just need to set `--onnx` as `True`.
* 2. cmake and make
@ -24,28 +25,32 @@ The Pytorch implementation is [DBNet](https://github.com/BaofengZan/DBNet.pytorc
cd build
cmake ..
make
cp /your_wts_path/DBNet.wts .
sudo ./dbnet -s // serialize model to plan file i.e. 'DBNet.engine'
sudo ./dbnet -d ../samples // deserialize plan file and run inference, the images in samples will be processed.
sudo ./dbnet -d ./test_imgs // deserialize plan file and run inference, all images in test_imgs folder will be processed.
```
## For windows
https://github.com/BaofengZan/DBNet-TensorRT
## Todo
* 1. ~~In common.hpp, the following two functions can be merged.~~
- [x] 1. In `common.hpp`, the following two functions can be merged.
```c++
ILayer* convBnLeaky(INetworkDefinition *network, std::map<std::string, Weights>& weightMap, ITensor& input, int outch, int ksize, int s, int g, std::string lname, bool bias = true)
```
```c++
ILayer* convBnLeaky(INetworkDefinition *network, std::map<std::string, Weights>& weightMap, ITensor& input, int outch, int ksize, int s, int g, std::string lname, bool bias = true)
```
```c++
ILayer* convBnLeaky2(INetworkDefinition *network, std::map<std::string, Weights>& weightMap, ITensor& input, int outch, int ksize, int s, int g, std::string lname, bool bias = true)
```
```c++
ILayer* convBnLeaky2(INetworkDefinition *network, std::map<std::string, Weights>& weightMap, ITensor& input, int outch, int ksize, int s, int g, std::string lname, bool bias = true)
```
* 2. The postprocess method here should be optimized, which is a little different from pytorch side.
- [x] 2. The postprocess method here should be optimized, which is a little different from pytorch side.
* 3. ~~The input image here is resized to 640x640 directly, while the pytorch side is using `letterbox` method.~~
- [x] 3. The input image here is resized to `640 x 640` directly, while the pytorch side is using `letterbox` method.