Run the original YOLO with Jetson Nano
What to do in this article
In this article, we will use the weights generated on colab to run on Jetson. Please refer to past articles for how to create an original model of YOLO. https://qiita.com/tayutayufk/items/4e5e35822edc5fda60ca https://qiita.com/tayutayufk/items/4dba4087e6f06fec338b
Jetson Nano ready
As a prerequisite, please install JetCard in Jetson.
First, download OpenCV.
https://qiita.com/usk81/items/98e54e2463e9d8a11415
Please refer to this site for installation.
I cloned and built under / home /" username "/ Lib /.
Next, we will introduce darknet.
Go to the directory where you want to put darknet
git clone https://github.com/AlexeyAB/darknet
After downloading, change the Makefile like colab before building.
GPU=1
CUDNN=1
CUDNN_HALF=0
OPENCV=1
AVX=0
OPENMP=0
LIBSO=1
ZED_CAMERA=0 # ZED SDK 3.0 and above
ZED_CAMERA_v2_8=0 # ZED SDK 2.X
# set GPU=1 and CUDNN=1 to speedup on GPU
# set CUDNN_HALF=1 to further speedup 3 x times (Mixed-precision on Tensor Cores) GPU: Volta, Xavier, Turing and higher
# set AVX=1 and OPENMP=1 to speedup on CPU (if error occurs then set AVX=0)
USE_CPP=0
DEBUG=0
ARCH= -gencode arch=compute_30,code=sm_30 \
-gencode arch=compute_35,code=sm_35 \
-gencode arch=compute_50,code=[sm_50,compute_50] \
-gencode arch=compute_52,code=[sm_52,compute_52] \
# -gencode arch=compute_61,code=[sm_61,compute_61]
OS := $(shell uname)
# Tesla V100
# ARCH= -gencode arch=compute_70,code=[sm_70,compute_70]
# GeForce RTX 2080 Ti, RTX 2080, RTX 2070, Quadro RTX 8000, Quadro RTX 6000, Quadro RTX 5000, Tesla T4, XNOR Tensor Cores
# ARCH= -gencode arch=compute_75,code=[sm_75,compute_75]
# Jetson XAVIER
# ARCH= -gencode arch=compute_72,code=[sm_72,compute_72]
# GTX 1080, GTX 1070, GTX 1060, GTX 1050, GTX 1030, Titan Xp, Tesla P40, Tesla P4
# ARCH= -gencode arch=compute_61,code=sm_61 -gencode arch=compute_61,code=compute_61
# GP100/Tesla P100 - DGX-1
# ARCH= -gencode arch=compute_60,code=sm_60
# For Jetson TX1, Tegra X1, DRIVE CX, DRIVE PX - uncomment:
ARCH= -gencode arch=compute_53,code=[sm_53,compute_53]
# For Jetson Tx2 or Drive-PX2 uncomment:
# ARCH= -gencode arch=compute_62,code=[sm_62,compute_62]
VPATH=./src/
EXEC=darknet
OBJDIR=./obj/
ifeq ($(LIBSO), 1)
LIBNAMESO=libdarknet.so
APPNAMESO=uselib
endif
ifeq ($(USE_CPP), 1)
CC=g++
else
CC=gcc
endif
CPP=g++ -std=c++11
NVCC=/usr/local/cuda/bin/nvcc
OPTS=-Ofast
LDFLAGS= -lm -pthread
COMMON= -Iinclude/ -I3rdparty/stb/include
CFLAGS=-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC
ifeq ($(DEBUG), 1)
#OPTS= -O0 -g
#OPTS= -Og -g
COMMON+= -DDEBUG
CFLAGS+= -DDEBUG
else ifeq ($(AVX), 1) CFLAGS+= -ffp-contract=fast -mavx -mavx2 -msse3 -msse4.1 -msse4.2 -msse4a endif endif
Change it to something like. Specifically at the top
GPU=0
CUDNN=0
CUDNN_HALF=0
OPENCV=0
AVX=0
OPENMP=0
LIBSO=0
Changed as follows
GPU=1
CUDNN=1
CUDNN_HALF=0
OPENCV=1
AVX=0
OPENMP=0
LIBSO=1
next
ARCH= -gencode arch=compute_30,code=sm_30 \
-gencode arch=compute_35,code=sm_35 \
-gencode arch=compute_50,code=[sm_50,compute_50] \
-gencode arch=compute_52,code=[sm_52,compute_52] \
-gencode arch=compute_61,code=[sm_61,compute_61]
.......................
# For Jetson TX1, Tegra X1, DRIVE CX, DRIVE PX - uncomment:
#ARCH= -gencode arch=compute_53,code=[sm_53,compute_53]
So I commented out the last line of -gencode, and since Jetson Nano uses the Jetson X1 chip, uncomment under# For Jetson TX1, Tegra X1, DRIVE CX, DRIVE PX --uncomment: Please give me.
Finally, I want to specify the location of jetson's nvcc, so I changed the location of NVCC
NVCC=/usr/local/cuda/bin/nvcc
change to
After that, make will start compiling.
Load darknet in python
Ehhhh
There is darknet_video.py in ./darknet/, so I will borrow it as much as I can.
darknet_video.py
from ctypes import *
import math
import random
import os
import cv2
import numpy as np
import time
import darknet
def convertBack(x, y, w, h):
xmin = int(round(x - (w / 2)))
xmax = int(round(x + (w / 2)))
ymin = int(round(y - (h / 2)))
ymax = int(round(y + (h / 2)))
return xmin, ymin, xmax, ymax
def cvDrawBoxes(detections, img):
for detection in detections:
x, y, w, h = detection[2][0],\
detection[2][1],\
detection[2][2],\
detection[2][3]
xmin, ymin, xmax, ymax = convertBack(
float(x), float(y), float(w), float(h))
pt1 = (xmin, ymin)
pt2 = (xmax, ymax)
cv2.rectangle(img, pt1, pt2, (0, 255, 0), 1)
cv2.putText(img,
detection[0].decode() +
" [" + str(round(detection[1] * 100, 2)) + "]",
(pt1[0], pt1[1] - 5), cv2.FONT_HERSHEY_SIMPLEX, 0.5,
[0, 255, 0], 2)
return img
netMain = None
metaMain = None
altNames = None
def YOLO():
global metaMain, netMain, altNames
configPath = "./cfg/yolov3-tiny.cfg"#Specify the location of the model cfg file here
weightPath = "./yolov3-tiny_final.weights"#Location of weights file
metaPath = "./cfg/obj.data"#data file location
if not os.path.exists(configPath):
raise ValueError("Invalid config path `" +
os.path.abspath(configPath)+"`")
if not os.path.exists(weightPath):
raise ValueError("Invalid weight path `" +
os.path.abspath(weightPath)+"`")
if not os.path.exists(metaPath):
raise ValueError("Invalid data file path `" +
os.path.abspath(metaPath)+"`")
if netMain is None:
netMain = darknet.load_net_custom(configPath.encode(
"ascii"), weightPath.encode("ascii"), 0, 1) # batch size = 1
if metaMain is None:
metaMain = darknet.load_meta(metaPath.encode("ascii"))
if altNames is None:
try:
with open(metaPath) as metaFH:
metaContents = metaFH.read()
import re
match = re.search("names *= *(.*)$", metaContents,
re.IGNORECASE | re.MULTILINE)
if match:
result = match.group(1)
else:
result = None
try:
if os.path.exists(result):
with open(result) as namesFH:
namesList = namesFH.read().strip().split("\n")
altNames = [x.strip() for x in namesList]
except TypeError:
pass
except Exception:
pass
cap = cv2.VideoCapture(0)#Uncomment here if you want to capture video from a webcam&Come out below
#cap = cv2.VideoCapture("test.mp4")
cap.set(3, 1280)
cap.set(4, 720)
out = cv2.VideoWriter(
"output.avi", cv2.VideoWriter_fourcc(*"MJPG"), 10.0,
(darknet.network_width(netMain), darknet.network_height(netMain)))
print("Starting the YOLO loop...")
darknet_image = darknet.make_image(darknet.network_width(netMain),
darknet.network_height(netMain),3)
while True:
prev_time = time.time()
ret, frame_read = cap.read()
frame_rgb = cv2.cvtColor(frame_read, cv2.COLOR_BGR2RGB)
frame_resized = cv2.resize(frame_rgb,
(darknet.network_width(netMain),
darknet.network_height(netMain)),
interpolation=cv2.INTER_LINEAR)
darknet.copy_image_from_bytes(darknet_image,frame_resized.tobytes())
detections = darknet.detect_image(netMain, metaMain, darknet_image, thresh=0.25)
#image = frame_resized
image = cvDrawBoxes(detections, frame_resized)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
print(1/(time.time()-prev_time))
cv2.imshow('Demo', image)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cap.release()
out.release()
if __name__ == "__main__":
YOLO()
It is not necessary to explain to those who usually use OpenCV with python. If you change the input file and select whether it is a webcam or a video file, it will work. What a wonderful thing. I wanted an end button, so towards the end
if cv2.waitKey(1)
To
if cv2.waitKey(1) & 0xFF == ord('q'):
break
Changed to
Run
Jetson is aware of Jetson. This shows that Jetson is awakened to self-consciousness and has the intelligence of a primate. (a big lie)
Well, as I still recognize the keyboard as myself, I feel that there are not enough samples. FPS was about 6 ~ 7fps. I want you to speed up a little. Is that the limit of python? Coding in C ++ ..... think
Finally
YOLO is good for people who want to recognize images with robots.
Recommended Posts