A implementation of yolov5 s/m//l/x with TensorRT API, it supports serialization/deserialization+image detection/video detection and recording.
This implementation bases on https://github.com/wang-xinyu/tensorrtx/tree/master/yolov5 but made improvement, such as get_rect() and add code for video detection and recording.
Usage:
mkdir build && cd build
cmake ..
make
./yolov5 -s [.wts] [.engine] [s/m/l/x or c gd gw]
./yolov5 -d [.engine] [sample_image_dir]
./yolov5 -v [.engine] video_file