tensorrt_demos：yolov3、yolov4转ONNX转TensorRT

最新推荐文章于 2024-08-10 08:03:21 发布

orzchenyuming

最新推荐文章于 2024-08-10 08:03:21 发布

阅读量1.1k

点赞数 2

分类专栏： onnx tensorrt 文章标签： tensorrt onnx Jetson tx2

本文链接：https://blog.youkuaiyun.com/qq_29936933/article/details/115250626

版权

onnx 同时被 2 个专栏收录

3 篇文章

订阅专栏

tensorrt

3 篇文章

订阅专栏

本文档详细介绍了如何在Ubuntu 18.0.4和Jetson TX2上配置TensorRT 7.1.3.4环境，支持YoloV3和YoloV4模型。步骤包括安装必要的库如OpenCV、CUDA、CUDNN，以及构建TensorRT演示项目。同时，提供了模型转换为ONNX和TensorRT引擎的步骤，最后展示了使用TensorRT运行YoloV4模型进行目标检测的例子。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

系统支持

实测在ubuntu18.0.4和Jetson 4.5.1 TX2可以运行。

yolov3、yolov4环境支持

TensorRT-7.1.3.4（实测用6.x.x.x也是可以的）
onnx==1.4.1 # 实测版本大于1.4.1时会报错
opencv4.1.1 #大于3.4的版本基本都可以
cuda10.2 (可能还要cudnn 8.x.x)

环境准备

Clone this repository.

$ cd ${HOME}/project
$ git clone https://github.com/jkjung-avt/tensorrt_demos.git
$ cd tensorrt_demos
$ sudo pip3 install Cython
$ cd ${HOME}/project/tensorrt_demos
$ make
$ cd ${HOME}/project/tensorrt_demos/plugins
$ make

yolov3、yolov4转模型

安装pycuda

$ cd ${HOME}/project/tensorrt_demos/ssd
$ ./install_pycuda.sh

安装onnx==1.4.1，大于这个版本会报错

$ sudo pip3 install onnx==1.4.1

Go to the “plugins/” subdirectory and build the “yolo_layer” plugin. When done, a “libyolo_layer.so” would be generated.

$ cd ${HOME}/project/tensorrt_demos/plugins
$ make

Download the pre-trained yolov3/yolov4 COCO models and convert the targeted model to ONNX and then to TensorRT engine. I use “yolov4-416” as example below. (Supported models: “yolov3-tiny-288”, “yolov3-tiny-416”, “yolov3-288”, “yolov3-416”, “yolov3-608”, “yolov3-spp-288”, “yolov3-spp-416”, “yolov3-spp-608”, “yolov4-tiny-288”, “yolov4-tiny-416”, “yolov4-288”, “yolov4-416”, “yolov4-608”, “yolov4-csp-256”, “yolov4-csp-512”, “yolov4x-mish-320”, “yolov4x-mish-640”, and custom models such as “yolov4-416x256”.)

$ cd ${HOME}/project/tensorrt_demos/yolo
$ ./download_yolo.sh
$ python3 yolo_to_onnx.py -m yolov4-416
$ python3 onnx_to_tensorrt.py -m yolov4-416

Test the TensorRT “yolov4-416” engine with the “dog.jpg” image.

$ cd ${HOME}/project/tensorrt_demos
$ wget https://raw.githubusercontent.com/pjreddie/darknet/master/data/dog.jpg -O ${HOME}/Pictures/dog.jpg
$ python3 trt_yolo.py --image ${HOME}/Pictures/dog.jpg \
                      -m yolov4-416
$ python3 trt_yolo.py --usb 1 --model yolov4-416 # 用usb摄像头
$ python3 trt_yolo.py --onboard 0 --model yolov4-416 # 用板卡上的摄像头，但是不一定成功