History

PeterH0323 b74db15d87 Optimize dbnet (#464 ) * 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		2021-04-07 17:54:27 +08:00
..
clipper	Optimize dbnet (#464 )	2021-04-07 17:54:27 +08:00
CMakeLists.txt	Optimize dbnet (#464 )	2021-04-07 17:54:27 +08:00
common.hpp	Optimize dbnet (#464 )	2021-04-07 17:54:27 +08:00
dbnet.cpp	Optimize dbnet (#464 )	2021-04-07 17:54:27 +08:00
logging.h	update readme	2020-11-06 22:54:15 +08:00
README.md	DBNet update (#318 )	2020-12-04 11:38:12 +08:00
utils.h	update readme	2020-11-06 22:54:15 +08:00

README.md

DBNet

The Pytorch implementation is DBNet.

How to Run

1. generate .wts
Download code and model from DBNet and config your environments.

In tools/predict.py, set save_wts as True, and run, the .wts will be generated.

onnx can also be exported, just need to set onnx as True.

cmake and make

mkdir build
cd build
cmake ..
make
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.

For windows

https://github.com/BaofengZan/DBNet-TensorRT

Todo

1. ~~In common.hpp, the following two functions can be merged.~~

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)

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)

1. The postprocess method here should be optimized, which is a little different from pytorch side.
1. ~~The input image here is resized to 640x640 directly, while the pytorch side is using letterbox method.~~