yolov8pose 部署版本，将DFL放在后处理中，部署rk3588.
1）编译
cd examples/rknn_yolov8pose_dfl_demo
bash build-linux_RK3588.sh
2）运行
cd install/rknn_yolov8pose_demo_Linux
./rknn_yolov8pose_demo
注意：修改模型、测试图像、保存图像的路径，修改文件为src下的main.cc
int main(int argc, char **argv)
{
char model_path[256] = "/home/firefly/zhangqian/rknn/examples/rknn_yolov8pose_dfl_demo/model/RK3588/yolov8n_pose.rknn";
char image_path[256] = "/home/firefly/zhangqian/rknn/examples/rknn_yolov8pose_dfl_demo/test.jpg";
char save_image_path[256] = "/home/firefly/zhangqian/rknn/examples/rknn_yolov8pose_dfl_demo/test_result.jpg";
detect(model_path, image_path, save_image_path);
return 0;
}
把板端模型推理和后处理时耗也附上，供参考，使用的芯片rk3588，模型输入640x640。
【yolov8pose 部署rknn（rk3588）、部署地平线Horizon、部署TensorRT，部署工程难度小、模型推理速度快，DFL放后处理中】