TX2入门教程软件篇-安装Yolo v3(jetpack4.2.3)

TX2入门教程软件篇-安装Yolo v3(jetpack4.2.3)

说明：

• 介绍在tx2下安装安装Yolo v3
• 环境：jetpack4.2.3 + CUDA10 + opencv 3.3.1

步骤：

• 先增加swap分区，要不后面运行测试容易出现内存不足导致失败，参考增加swap教程
• 下载darknet

mkdir -p ~/dl/darknet
cd ~/dl/darknet
git clone https://github.com/pjreddie/darknet.git
cd darknet

• 修改Makefile文件，启用GPU和opencv
• CUDA 架构 TX2 是“62”, TX1 “53”

GPU=1
CUDNN=1
OPENCV=1
OPENMP=0
DEBUG=0
 
ARCH= -gencode arch=compute_53,code=[sm_53,compute_53] \
      -gencode arch=compute_62,code=[sm_62,compute_62]
#      -gencode arch=compute_20,code=[sm_20,sm_21] \ This one is deprecated?
# This is what I use, uncomment if you know your arch and want to specify
# ARCH= -gencode arch=compute_52,code=compute_52

• 编译

make -j4

• 判断成功

./darknet 
usage: ./darknet <function>

测试yolov3

• 下载训练好的yolov3模型

wget https://pjreddie.com/media/files/yolov3.weights

• 测试运行

./darknet detect cfg/yolov3.cfg yolov3.weights data/dog.jpg

• 下载训练好的tiny yolov3模型

wget https://pjreddie.com/media/files/yolov3-tiny.weights

• 测试运行

./darknet detect cfg/yolov3-tiny.cfg yolov3-tiny.weights data/dog.jpg

• 使用摄像头进行实时检测
• 默认tx2的内置摄像头占用了video0，使用-c指定video1

./darknet detector demo cfg/coco.data cfg/yolov3.cfg yolov3.weights -c 1 

• 或使用视频文件

./darknet detector demo cfg/coco.data cfg/yolov3.cfg yolov3.weights <video file>

• 或使用板载摄像头

./darknet detector demo cfg/coco.data cfg/yolov3.cfg yolov3.weights "nvcamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height(int)720, format=(string)I420, framerate=(fraction)30/1 ! nvvidconv flip-method=0 ! video/x-raw, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink"

参考：

• https://pjreddie.com/darknet/yolo/
• https://pjreddie.com/darknet/install/