Yolo v3 Object Detection - How to Implement Yolo V3 Object Detection on Windows with GPU | Tut #1

Want to Learn YOLOv7 and solve real-world problems?
====================

Struggling to implement real-time Yolo V3 on a GPU? Well, just watch this Computer Vision Tutorial to learn how quick and easy it is to implement Yolo V3 Object Detection using PyTorch on Windows 11. We use OpenCV Python for our implementation:

===This Video is Sponsored by Altium=======
======================================

[FREE Webinar] Top 10 Secrets to Develop AI apps in Computer Vision

Hey guys and welcome back, so in this video I'm going to show you how to implement Yolo V3 Object Detection using PyTorch on Windows 10. We'll walk through everything from requirements to setup, then all the way to executing the CNN. Now we are not going to go through the theory as I mentioned earlier, this will just be hands-on practical tutorial to get you up and running.

Just be aware that I have not been able to train YoloV3 on Windows, but this turned out to be a blessing in disguise as I now use a new workflow called Supervisely. I will show you how we can use this new workflow for annotation, to training as well as data augmentation. But for now lets get started with the execution of Yolo V3

If you want to skip this video of image scraping and annotation, you can go over the github repo, and download the annotated dataset.
------------------------------------------------------------
Support us on Patreon
Chat to us on Discord
Interact with us on Facebook
Check my latest work on Instagram
Learn Advanced Tutorials on Udemy
------------------------------------------------------------
To learn more on Artificial Intelligence, IoT, Deep Learning FPGAs, Arduinos, PCB Design and Image Processing then check out

Please like and Subscribe for more videos :)

[varadpadalkar]

i got one error in anaconda prompt :
Solving environment: failed

ResolvePackageNotFound:
- vs2015_runtime==15.5.2=3
- vc==14.1=h0510ff6_3
- make==3.82=0
what should i do? please help

[mezosawy]

you are the best, Thank you so much for this impressive tutorial.

[hasithaedirisinghe]

kivy 1.10.1 requires Kivy-Garden>=0.1.4, which is not installed.
tensorflow 1.9.0 has requirement setuptools<=39.1.0, but you'll have setuptools 40.0.0 which is incompatible.
these two errors prompted after typing conda env create -f yolo.yml

[anhphuong]

Sir! I have this error when I used "conda env create -f yolo.yml" it pop up
"Solving environment: failed

ResolvePackageNotFound:
- vs2015_runtime==15.5.2=3
- make==3.82=0
- vc==14.1=h0510ff6_3"

Please help me fix this!!!

[hglide]

Thank you for the great tutorial video!

[zaferatakan]

Hi.. Thank you for the video but I'm keep getting CuDnn error : CUDNN_STATUS_SUCCESS .. I tried to install cudnn again via anaconda prompt but still the same.. do you have any suggestions ? Thanks

[nicknameezi]

ResolvePackageNotFound:
- vc==14.1=h0510ff6_3
- make==3.82=0
- vs2015_runtime==15.5.2=3
I installed VS2019, and Anaconda Python 3.7 on Windows 10 x64
On "conda env create -f yolo.yml",
Why do I get these errors? Edit: It solved when I removed the version numbers (at yolo.yml file 3 lines).

[alphaquanta]

Hi there, first of all great work. I just have few questions;
1 - How I can / Can I set desired amount of cores ?
2 - How can I see if the program runs on GPU or CPU, is there any indication or configration.


Questions might be silly, pardon me ^^

[calmato]

I followed every step, i got a error:
import torch
ModuleNotFoundError: no module named torch

[truongpham]

i cant used "conda env create -f yolo.yml " solving environment: failed
ResolvePackageNotFound:
- vs2015_runtime==15.5.2=3
- make==3.82=0
- vc==14.1=h0510ff6_3
help me please!!

[bhargavvarma]

Hi I have an issue after running the video file. It is showing the error as

UserWarning: nn.Upsampling is deprecated. Use nn.functional.interpolate instead.
warnings.warn("nn.Upsampling is deprecated. Use nn.functional.interpolate instead.")
warning: Error opening file
Traceback (most recent call last):
File "video_demo.py", line 119, in <module>
assert cap.isOpened(), 'Cannot capture source'
AssertionError: Cannot capture source

can you help me with it.

[farang]

i am at 0.3 fps in video, i have GTX 1070. i don't know how to make usage of gpu. edit : i found my error, i was in a virtualenv, don't use venv it doesnt work, in fresh install it worked with gpu

[rayomondchinoy]

Excellent tutorial. Really appreciate you spending time on this. I already have cuda 10.2 installed, as I need it for other purposes. will your process work on it? I don't have CUDNN installed yet either

[roh]

how to print the number of objects in the frame.

[hanifeguney]

when executing the yolo.yml file with environment it gives me error:
Solving environment: failed ResolvePackageNotFound:
- make==3.82=0
- vc==14.1=h0510ff6_3
- vs2015_runtime==15.5.2=3
I changed the last 2 files' vesion in yolo.yml file and remove the make==3.82=0 line and it worked! I hope everything will works completely.

[tibialfoldi]

can someone send me the github repository for yolo3workflow..?

[vikramreddy]

Thanks so much I jus entered in to this topic worked along. With you as in video got the result jus like you but FPS of 12 any how thanks for your efforts

[febryrh]

hi, i trying step that u explain on this video, but i have some problem. this is my problem

(base) C:\yolo>conda activate yolov3
Could not find conda environment: yolov3
You can list all discoverable environments with `conda info --envs`.


can u help me solve this problem?

[emmadaily]

at 2:59, can you tell me where "conda" comes from and when i did step 3, there were some errors as well as "cannot find path specified", how can fix it?

[marieandrefaye]

how to download cuDNN v7.5 ... because I don't have an NVDIA account please???,