tensorRT for Yolov3

Ubuntu  16.04  or Jetson Nano
TensorRT 5.1/5.0.2.6/4.0.1.6
CUDA 9.2 or CUDA 9.0 or CUDA 10.0

Download the caffe model converted by official model:

• Baidu Cloud here pwd: gbue
• Google Drive here

If run model trained by yourself, comment the "upsample_param" blocks, and modify the prototxt the last layer as:

layer {
    #the bottoms are the yolo input layers
    bottom: "layer82-conv"
    bottom: "layer94-conv"
    bottom: "layer106-conv"
    top: "yolo-det"
    name: "yolo-det"
    type: "Yolo"
}

It also needs to change the yolo configs in "YoloConfigs.h" if different kernels.

git submodule update --init --recursive mkdir build cd build && cmake .. && make && make install cd ..

1.Added multithreading

2.Added tag name

3.Added video inference

./install/runYolov3 --caffemodel=./yolov3_608.caffemodel --prototxt=./yolov3_608.prototxt --display=1 --inputstream=video --videofile=sample_720p.mp4 --classname=coco.names

./install/runYolov3 --caffemodel=./yolov3_608.caffemodel --prototxt=./yolov3_608.prototxt --display=1 --inputstream=cam --cam=0 --classname=coco.names

./install/runYolov3 --caffemodel=./yolov3_608.caffemodel --prototxt=./yolov3_608.prototxt --display=1 --inputstream=cam --cam=0 --classname=coco.names --mode=int8 --calib=cal.list