Object Detection Web Application with Flask and YOLOv9

Learn how to create your own Object Detection Web Application using Flask and YOLOv9. In this tutorial, I'll walk through the process step by step, empowering you to deploy your own powerful AI-driven tool for object recognition on the web.
GitHub: github.com/AarohiSingla/Objec...
For queries: You can comment in comment section or you can email me at [e-mail]
#computervision #yolov9 #objectdetection #yolov8 #flask #webapplicationdevelopment

[@Sunil-ez1hx]

Your channel is the best for Artificial Intelligence & Data Science upto date knowledge with latest topics

[@CodeWithAarohi]

Glad you think so!

[@kevingiftson9420]

Thank you,you're a life saver

[@CodeWithAarohi]

Happy to help

[@arnavthakur5409]

Thank you ma'am for making this useful video. Your channel is really helping me building my career in Artificial Intelligence & many of your topics are related to projects our company is working upon. Thank you again & keep sharing your videos

[@CodeWithAarohi]

It's my pleasure

[@GuanLaoban]

Thanks, Aarohi..! Great videos!

[@CodeWithAarohi]

Thank you! Glad my video is helpful 🙂

[@soravsingla8782]

What an amazing video👏👏👏

[@CodeWithAarohi]

Thank you so much

[@amanullah1583]

excellent video

[@CodeWithAarohi]

Thanks!

[@bayda3695]

thanks a lot, when I run the code the results video not embedded in the same site it displayed in a separate window. why?

[@SerSer-xi6et]

thank you very much for the quality content. It’s very interesting to try, but tell me why the code located on Github does not have a display function when launching webcams and starting an RTSP stream. (or maybe it didn’t start for me.) My native language is not English, so the text may be a little incorrect.

[@sreethasetty]

great video!!

[@CodeWithAarohi]

Thank you!!

[@Vnobile]

for me i couldnt use the yolov9 weight as it says it does not match ultralytics model which is only yolov8

[@hieuhoanghuu35]

why I clone code on your git have not button Start webcam and Start RSTP Stream

[@CodeWithAarohi]

I have mentioned in last few minutes of my video that there are 2 versions of this code. 1st version (where you can test on images and videos) is available for all. 2nd version where you can get entire code is for Channel members (Contribution Level 2).

[@waseemhaider5295]

great work ,and i want to create project to select one person according to user choice in one camera and detect our system exact person in any camera or video and track successfully, please mam one video for this type of project , love from Pakistan

[@CodeWithAarohi]

Step 1: Create an object detection model to detect all persons.
Step 2: Compare the features of the selected person with the features of each detected person to identify the exact person chosen by the user.
Extract features from both the selected person and all detected persons in the camera feed.
You can use Histogram of Oriented Gradients (HOG), deep learning embeddings (e.g., from convolutional neural networks)
Once you have extracted features for the selected person and all detected persons, you can compare these features to find the best match.
This comparison can be done using various distance metrics such as Euclidean distance, cosine similarity etc.
Set a threshold to determine the best match between the selected person and the detected persons.

[@mghirbirahma4733]

great video.. can i work on collab and vs code ??

[@sanjay6242]

how to run custom model / Weights ?

[@pifordtechnologiespvtltd5698]

Nice

[@CodeWithAarohi]

Thanks

[@umutcelik9821]

nice job Aarohi

[@CodeWithAarohi]

Thanks!

[@busra1494]

Great project but I have a question. When I run the project, I always want to see the prediction screen on the Yolov9 Object Detection Result page. How can I solve it?

[@CodeWithAarohi]

You need to make minor modifications in your function in order to redirect.

[@TomislavLevakovic]

Hi ! Great channel ! Lot to learn ! Can ypu explain more precisely where to download full code for this video i have just start Contribution Level 2 :-)

[@CodeWithAarohi]

Mail me at [e-mail]
I will send you code there.

[@omerjadoon]

Thanks for the video and code. But i couldnt find any code related to web cam in your github repo.

[@CodeWithAarohi]

It is because that code code is available for channel members (Contribution level 2)

[@saraalkhalily6400]

hi , if i want to display the image in HTML page instead of video how can i do ?

[@CodeWithAarohi]

You need to make changes in your webapp.py file. There, we have a function which is responsible for showing the output after detection. You can modify that function.

[@AmandeepSingh-uq3wp]

How did you get the static folder? I am not able to understand its code.

[@CodeWithAarohi]

Static folder have files related to css and javascript. These files are used to design the dashboard. I picked these files from internet. No need to understand these files. They are only helpful to design the app.

[@VTVoice761]

Mam which version of tensorflow is required

[@CodeWithAarohi]

You can just run pip install tensorflow. It will work with the latest one also

[@akhilsajeevan1192]

Hi, I have a doubt.
I trained a yolov8 model with a 4 claases with my custom dataset. Now i want to add a new 2 classes into that model with another dataset but i dont want to loss the older weights of the best model when trained , which is already trained with 4 classes.
Is there any solutions or reference to resolve this?
Is freezing the model layers may be helpful?
Is that so can you show me to resolve it

[@CodeWithAarohi]

Yes, you can add new classes to your model without losing the weights already learned for the existing classes. Use transfer learning (freezing some layers of the pre-trained model) to retain their learned features while allowing new layers to adapt to the new classes.

[@akhilsajeevan1192]

@@CodeWithAarohi Can you show any documentation or links for doing this procedure .
step by step simple method

[@poojakabirdas9006]

Thanks Aarohi, good content, meanwhile I have a doubt, how to access this web app from a remote location.

[@CodeWithAarohi]

You can host it on any server and then you can access it from anywhere.

[@AbhishekKumar-cd7co]

Hello mam, how to get the full code because i am facing error while performing with this code.

[@CodeWithAarohi]

This code will work perfectly for image and video detection. Share error.

[@leejaric4448]

Hi madam, is it possible to make an integration between the yolo v8 output with react native app, so i can view the webcam output and the data on my app, you have any idea on this?

[@CodeWithAarohi]

Sorry, No idea

[@leejaric4448]

@@CodeWithAarohi Madam, Can you show me the code of the webcam, I really need it T_T

[@CodeWithAarohi]

@@leejaric4448 As I have mentioned in the end of this video that code for image and video is available for all users but channel members ( Contribution level 2) can access the code of webcam and rtsp stream also.

[@bijayaojha8288]

mam please upload for licence plate detection also

[@CodeWithAarohi]

The code will be same. You just need to change the name of trained model in your webapp.py file.

[@romeroe5931]

How to create yolov9 custom detection web application with flask?

[@CodeWithAarohi]

This is what I have explained in this video. Train your custom model using yolov9. Then provide the path of that custom model in web_app.py file.

[@mghirbirahma4733]

how can i be from your channel member plz ??

[@CodeWithAarohi]

Click on this link and then click on join button. Follow on screen instructions. ua-cam.com/channels/gHDngFV50KmbqF_6-K8XhA.html

[@sohanwagh5447]

can i pass my custom yolo v9 model

[@CodeWithAarohi]

Yes

[@user-cs9en8wj5g]

The title of the video is "Object Detection Web Application with Flask and YOLOv9", but you're using YOLOv8. Please fix the title.

[@CodeWithAarohi]

Sure! Thanks