ฝัง
- เผยแพร่เมื่อ 27 ก.ย. 2024
- We’re going to learn in this tutorial how to install and run Yolo on the Nvidia Jetson Nano using its 128 cuda cores gpu.
Instructions: pysource.com/2...
➤ Full Videocourses:
Object Detection: pysource.com/o...
➤ Follow me on:
Instagram: / pysource7
LinkedIn: / pysource
➤ For business inquiries:
pysource.com/c...
![[🔴LIVE] งานประกาศรางวัล MAYA TV AWARDS 2024 : มายามหาชน](http://i.ytimg.com/vi/nwmWYXXVwBY/mqdefault.jpg)
You saved my life! My problem was cuda declaration. Exporting the address solved that. Thank you, sir.
Thanks for sharing, at least we know that we need more gpu power for this kind of application
Wow, nareszcie polski filmik o Jetson Nano :)
Thank you for sharing, especially GPU inference performance demo.
I have followed every step with success but running the program with the code at 15:14 opens a screen of just green. Someone else has also commented this issue. Thanks for any help
I saw a comment down bellow that you are using the Jetson CPU rather than GPU. I think that is correct. 15 FPS should give a quite descent video and I saw no difference between the 2 FPS and the 15 FPS on the tiny Yolo.
The software he uses to share the screen of his jetson nano with his desktop limits the framerate.
nah there IS a difference
Please help on 7:54 when I use the make command, I get an error saying opencv2/core/version.hpp no such file or directory.
You are the best.
Thank you for the vidéo ! Which Jetson Nano Image did you flash ?
Hi! Nice tutorial, but i'm getting a error detecting with camera
terminal keeps looping this message
VIDIOC_QBUF: Invalid argument
Hi! Thankyou so much for the detailed tutorial! Could you kindly tell me how do I save the processed video feeds ?
Great! I check that Jetson nano has not enough power to demo..
what if I am using yolov5, Nvidia tx1
hay how did you install anydesk in jetson nano
In the output of the forward, you only use index 0-3 for the location, and index 5 to the end for the confidence score. then what is the function of the 4th index?
Would be great to see yolo running at 30fps on jetson nano
Yeah that would be great. Is there any way to upgrade the Fps on intel or jetnano?
There's just not much cuda cores just to run Yolo at 30fps compared to desktop GPU.
Getting error while running the yolov3 in jetson nano for live webcam- Gstreamer Warning - pipeline have not been created in opencv.Please help
Hello Sir,
Do you think Jetson Nano 4GB, is not the good choice to do project like self-driving car, because the Web Camera is too slow?
I get only 1.5 fps
Can you help me please?
Hi....I am running yolo3 in jetson nano using cuda 10.2. Getting the FPS of 0.3 which is very very slow...how to improve FPS?
would TensorRT helps for the low fps?
I have trouble to install OpenCv in my jetson guys. Can you please help me guys ?
what is the background musick?
What is the command for CSI camera instead of webcam ? Thanks
You should first install the CSI camera to work with Opencv and then to run it I recommend to follow the instructions here:
github.com/JetsonHacksNano/CSI-Camera
Hi, thanks for the video, could you tell me your configuration to access to your Jetson nano from VNC or another remote access program?
I am acces with NoMachine, and work.
Your Channel and tutorials are so cool....Keep up.... are you interested in publishing you work with significant modification if the model can be run in mobile (android .apk)....?
Hi, I'm going to focus at the moment mostly on Opencv wiht Python.
It's possible that in the future I will implement also Android as the computer vision apps are getting more and more popular also on the mobile side.
I have tried yolov3, but failed with out of memory error
Please deep learning tutorial
Should I install cuda tool kit before install Yolo?
I'm having an error which is "obj/convolutional_kernels.o' failed" in the end of "make" compiling (7:44)
okay I fixed the problem. If someone who's had same problem the error I mentioned, follow the steps below
At 3:11, when you export, instead of this...
export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
try this one
export PATH=/usr/local/cuda/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
then open the terminal, and go to darknet directory.
Do "make" command on the terminal at darknet directory
thanks bro, you solved my problem
@@beomseokpark1500 thanks bro, problem solved
@@beomseokpark1500 Thank you! Problem solved
try to use with darknet framework
Hello, Have you worked with YOLO v5 and tensorRT on a Jetson nano board? I am facing lots of issues while installing and making it work.
Would be a great help if you could guide.
Same error here... You solved it?
Hello sir, are we going to do these setups for .jetson xavier too? @Pysource
Hi, I will bring Jetson Xavier NX as well into the channel, as It's an interesting device, much better than the nano in terms of performances
@@pysource-com Yes, sir. I want to use it because of its performance. I will use LPRnet with my friend in my project. We made the license plate recognition system with tensorflow and cpu on windows with faster r-cnn inception v2 model. But we can't install any application in jetson nano or xavier. I don't know how we can handle this problem. We found a mentor to teach us. But his knowledge was not very good.
If you know about this, can you help? Can you mentor us hourly?
@@ozlematz2528 you can contact me through my website pysource.com
Hi, thanks for the great tutorials. However, I'm having an issue trying to use my iPhone as an ip webcam. So I downloaded an app called IP Webcam and used it which has its own IP address. So I type the command accordingly and when I run it "Stream end prematurely at 2428, should be 184467440737091615" message is printed followed by an endless loop of "Video-stream stopped!" error. Can you help me out if you can. I'd really appreciate it!
I cant evenrun the first prompt "sudo apt-get update" says sudu is not recognized as internal or external command.
Can you help me out
sorry but fps calculation seems wrong. 15-16 fps doesn't seem like that. It must be fluent.
The recording is not fluent because I'm using the Jetson Nano remotely from my computer, that's the reason why we see it this way, but in reality it looks fluent
@@pysource-com Thank you for your response. Your videos are really good.
Hi Sergio, everything works perfectly, thank you very much!
How can I do to work with an IP camera with RTSP? in real time, without delay.
Please support me, thanks.
Regards
Rudy
I cant evenrun the first prompt "sudo apt-get update" says sudu is not recognized as internal or external command.
Can you help me ou
Is sudo instead sudu
when you have error 1 after executing 'make' with include use this command:
sudo apt install nvidia-jetpack
export PATH=/usr/local/cuda/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
and then repeat 'make' in darknet folder
bro you are working hard but people are less on your channel.you should do some advertisment of your channel/website. 💕💖💖
i run into /darknet: No such file or directory at 11:19. please lmk if anyone solved this
I have this problem too were you able to solve it?
Hi sir. Can you help to make a drone detection system using jetson nano and yolov3 Or yolov8
Thank you for sharing, can you make video to install and run on Nvidia Jetson Nano with GPU yolov5 and yolov6?
Can you make a video about running Yolov3 or Yolov4 with TensorRT ?
I think that u will get more FPS by using DetecNet instead of Darknet
Edit : GPU of Jetson nano isnt too efficient to run YOLOv3 even with TensorRT optimisations... but with Jetson NX Xavier or AGX Xavier we can run YOLOv3 with a high FPS by using ONNX to parse to TensorRT or using the TF-TRT api
Btw DetecNet is an Nvidia DNN and include only CNN's (like ssd mobilenet ssd inception ...)which r mentionned on dusty-nv Hello AI world github and YOLO isn't included in DetectNet.
thank you for making this tutorial, may i ask, should we install cuda or not?thank you for making this tutorial, may i ask, should we install cuda or not?thank you for making this tutorial, may i ask, should we install cuda or not?thank you for making this tutorial, may i ask, should we install cuda or not?
too slow
hello , how can i add object tracking using 2 servomotors
Please could you do a video on how to code OpenCV to use GPU of Jetson Nano using Python.
Did you find how it's done?
Sir, Do you recommend 2gb Nvdia Jetson nano for startup of Deep learning in 2021?
Thank you for posting Jetson Nano videos
im having trouble downloading the weights itself..its taking a hell lot of time to download around 240 mb .. its showing eta as 24w 6h .. i have 50mbps connection.. whats going wrong here...? pls anyone help!!!!!!!!
Unfortunately I noticed there is a bandwith limit when downloading weights from darknet, no matter how fast your connection is, it will be slow
@@pysource-com thank you for replying.
Aren't there any work around for that ., 😅
Thanks bro, it works perfectly!
@Pysource how to I add my own car image to detect..because when I run, not detect car ..
can this be used with Intel Realsense d345 camera?
slow fps
Good video
I have a problem Demo needs OpenCV for webcam images.
Probably you don't have opencv installed.
Check this other tutorial to see how to install Opencv th-cam.com/video/SSzZZ68jy_I/w-d-xo.html
cd darknet/
gedit Makefile
change opencv 0 to 1
then go to the terminal and do (make)
you need to activate the opencv
I think you are using cpu instead of gpu
Hello, thanks for the tutorial ! I just want to know how to display the jetson nano ubuntu Graphical user interface on my windows PC. I am using PuTTY to ssh the jetson nano board from PC. Thank you in advance !
I am using NoMachine for acces from windows