Auto Annotation for generating segmentation dataset using YOLOv8 & SAM

[@Sunil-ez1hx]

What a great video ma’am. Very good presentation skills. Thanks

[@kadapallanithin]

Great work! Thanks for teaching for free

[@CodeWithAarohi]

You are welcome 🙂

[@shamarulo_9847]

Hello,. Awesome video. so simple.

[@CodeWithAarohi]

Hey, thanks!

[@soravsingla8782]

Hi Aarohi, your content is excellent and your channel is one of the best Artificial Intelligence channel but still not getting that much of likes which your channel deserves. Hope you succeed #AI #ArtificialIntelligence #DataScience #EducationalContent

[@CodeWithAarohi]

Thank you so much for your kind words and support! It means a lot to me. 😊🙏

[@karishmathumu5257]

Hello Arohi, This video is great. Wanted to ask, if this implementation can be done with yolov4 in combination with SAM ? Thank you.

[@CodeWithAarohi]

Yes, You can do that. You only need yolo model for detection of objects and once you have those detections (bboxes) , you can pass those bboxes to SAM.

[@raj_5117]

amazing, please also make some tutorial video on 3D point segmentation.

[@CodeWithAarohi]

I will try

[@raj_5117]

@@CodeWithAarohi Thank you so much.

[@olanrewajuatanda533]

Please, how do I serve multiple images and or image sequence as input and how do I save the output?

[@bharathprabakaran2229]

mam is it possible to use ultralytics in plant disease detection

[@CodeWithAarohi]

@@bharathprabakaran2229 yes, you can use

[@nouhailamokhtatif1060]

Hello, thank you for the vide, i want to know the nature of the output generated by sam, the first index is the object class and what about the other numbers?

[@AmosKibet-b3r]

How do you run the model in Colab with a camera?

[@ZeynepC3]

love this woman

[@Glam-by-Amna]

Hi Arohi. I wanted to ask if we can use this same method for medical images like Brain MRI's??

[@CodeWithAarohi]

Try to test your MRI images with SAM. If the class you want to segment is getting segmented by SAM then you can use this code for Brain MRI dataset.

[@Glam-by-Amna]

@@CodeWithAarohi ok thank you.

[@karlm9584]

This is great and thank you so much! Note that the detection model used must be a YOLOv8 detection model. I attempted to use a YOLOv5 model and it would not work. I am still training the YOLOv8 model on my dataset (5 days!), and after it is completed I will attempt to use it to perform the auto annotation. Thankfully the annotation format for YOLOv8 and YOLOv5 is the same for my 23,000 images, but the data.yaml is slightly different. (I don't want to annotate 23,000 images for segmentation manually, it already took me 3 months to annotate them just with bounding boxes!)

[@CodeWithAarohi]

I'm glad to hear the information was helpful! Training models on such large datasets can indeed be time-consuming, but it sounds like you're making solid progress. It's also a relief🙂 that the annotation format is the same for YOLOv8 and YOLOv5 for your dataset of 23,000 images.

[@16kthakrar]

How do you think this performs relative to using them independently?

[@prachirawat6963]

does this work on custom dataset as well? images that might not be available/ trained on theby pretrained object detection model?

[@CodeWithAarohi]

You need to train yolov8 on your custom dataset first and then use SAM on custom detections.

[@prachirawat6963]

@@CodeWithAarohi okay Thank you

[@johnanthonyvaldez764]

@@CodeWithAarohi how will that be possible? Do you have any video tutorials for that? I hope you do I badly need it for my project :(

[@Sri22752]

Can we use for this labelme tool

[@akhilsajeevan1192]

how to get class names or how to auto annotate custom class names on our custom dataset. Is there any option to get class names which the model segmented in the txt file or any other file what the model predicted ?

[@bouchakourlallouani8044]

How to use the yolo for texte detection

[@ashirahmed5600]

Hi, how can i extract these segments from the image and make new images with just the segment

[@deemon101]

Can you do another one of these with Yolov8 or 9 or 10 and SAM2?

[@CodeWithAarohi]

Sure!

[@nguyenviethoan1947]

thank you for your tutorial, it is very nice. Can I ask a question ? The segmentation labels are saved as txt. If I want to edit the auto-annotation segmentation, which labeltool can recognize these annotated txt files ?

[@shinwarikhan4677]

hello man can you make video on that what is anomoly detection, how we prepaere dataset for it and how make it on custom dataset.. we will be thankful

[@CodeWithAarohi]

Will try after finishing my pipelined work.

[@jeffreyeiyike2358]

Good day, I would like to extract yolo feature. What is the best way to do it

[@NoopurSrivastava018]

Will this procedure work on Satellite datasets as well? Since COCO doesn't have class information with respect to Satellite imagery.

[@CodeWithAarohi]

Yes

[@asfandiyar5829]

Thank you for video. The git repo has become dated unfortunately and I had to do a little bit of work to get this working. I'll try to merge my work later. Thanks again.

[@CodeWithAarohi]

Thanks!

[@reebalnofal2531]

Thanks a lot for your videos, i was wondering how can i use SAM if i already have the bounding boxes labels as [class x1 y1 x2 y2], should i use the last part of your code in this case?

[@CodeWithAarohi]

You can follow this video: kzbin.info/www/bejne/jnOcq5pvbt1loKc

[@reebalnofal2531]

@@CodeWithAarohi i can't fix this problem: "ImportError: Cannot load backend 'TkAgg' which requires the 'tk' interactive framework, as 'headless' is currently running"

[@ZhCrypto]

What does mean by image segmentation and where is code

[@NitishRaj]

HI Aarohi, Can we detect object using SAM generated labels file.

[@CodeWithAarohi]

SAM is an Image segmentation model. It will provide you masks but yes for each mask it has a box. You can get that box coordinates around the object.

[@adnanahmad4168]

I have trained custom Yolov8 model. How to deploy that model on jetson nano? Please make a video on it. I have tu submit my project within a week and i am stuck

[@CodeWithAarohi]

I will try

[@adnanahmad4168]

@@CodeWithAarohi Thankyou for replying. The custom model is trained for object detection on multiple classes in google collab. Please try to cover from that perspective.

[@zaidahmed4069]

does this annotation model producing X,Y coordinates values ?? if yes then where are they? please reply with a screenshot

[@Anuuuuuuii]

i have large collection of images of employees in a company. my task is label the employees sitting and standing their status, how to automatically label the large image data set

[@CodeWithAarohi]

If your task is to prepare image segmentation dataset then you can use this directly. If your task is to prepare object detection dataset then you can fetch the segmented mask using the auto labeling and then write a code which will put bounding box around the segmented mask.

[@ZeeshanAli-cl3tb]

I am using yolo for detecting the fire I have dataset of 10000 images how could I annotate all of the images

[@m.abdullahgamer4494]

HI. I am implementing your code. I have 5 images in the "images" folder. The jupyter shows inference that the results are saved in results/segment/predict. But the directory does not show results folder. The images folder and the jupyter file are in the same directory.plz guide

[@bugratasdemir6833]

Hi Aarohi. I want to detect all objects in an image and cut-out them with using SAM. If you help me. you make me happy. Thank you. Greetings from Turkey.

[@CodeWithAarohi]

Mail me at [e-mail]

[@bugratasdemir6833]

@@CodeWithAarohi I sent email to you. Thank you.

[@bugratasdemir6833]

@@CodeWithAarohi I sent email to you. Thank you.

[@Virendrakumar1]

How to get the SAM mask boundary box like YOLO

[@CodeWithAarohi]

There is a function show_box in SAM you can use it. Check this video: kzbin.info/www/bejne/jnOcq5pvbt1loKc In this I have explained that

[@rama_bhuyan]

Hi Thank you for the great explanation. I got a question- how do we get the bounding boxes for costume dataset ( class not in COCO/IMAGENet). Does the yolo model has to be trained/ if yes then how many minimum manual annotations required so that we could use this tool effectively/efficiently.

[@CodeWithAarohi]

You have to train yolo on custom dataset. Try with at least 500 images per class.

[@davisonyeoguzoro9232]

I have the same question. So I am to use my custom yolo model (best.pt), then use it with SAM for the auto annotation?

[@mpfmax0]

how can I actually plot the masks on the images and save them? I have the text files with the annotations but plot them to visually evaluate the resulting masks and various images and automatically save those plotted masks into a folder, kind of how you can do it with the bounding boxes from YOLOv8.

[@CodeWithAarohi]

Check this video: kzbin.info/www/bejne/jnOcq5pvbt1loKc

[@rangurangu3745]

Kognic annotation 2d images how to use this code auto select segmentation on running kognic web task...can you please tell me.

[@vikashkumar-cr7ee]

Hi Arrrohi, As I can see, SAM is annotating everything which is available in one frame, and all of the segmented parts is not of one's interest. So, how to get rid of this, or how to get selective segmentation?

[@CodeWithAarohi]

Check this video: kzbin.info/www/bejne/jnOcq5pvbt1loKc

[@amitpatel3071]

hey how do i get the code please ?

[@CodeWithAarohi]

github.com/AarohiSingla/Auto-Annotation-Using-YOLOv8-and-SAm

[@SVajrapuPrathyusha]

Hi mam... I am following your channel. Your videos are amazing🤩 I watched your traffic sign detection using yolo v4 and doing the project on that... I have some doubts in the code.. I posted the comments in that video. Please give your answer to my question which i posted their Thank you

[@CodeWithAarohi]

sure

[@kumargaurav2170]

But isn't it's counter intuitive because if we are able to build obj det model on custom dataset then that is more than enough for our SAM model. Why would we even need to prepare data for seg task as we can get our result from Sam model by just passing obj det output to it.

[@CodeWithAarohi]

Suppose you want to use any other segmentation model then you need Dataset. And preparing that dataset will be a time consuming process. You can annotate the dataset using SAM and object detection model then. But if you are using SAM as a segmentation model then there is no need to create the dataset.

[@olanrewajuatanda533]

@@CodeWithAarohi please, can I use this to prepare my dataset for object detection?

[@SuvramalyaBasak]

What i don't get is why use a segmentation model for generating segmentation annonation for training another custom segmentation model. If SAM can provide proper segmentations, why not just use it to obtain the segmentations. Why then try another segmentation model?

[@CodeWithAarohi]

SAM provide segmentations but not labels for those segmentations. We are getting class labels using yolov8 object detections and then performing segmentation only on the objects of our choice.

[@Omsip123]

Please do not finish every sentence with OK....❤

[@CodeWithAarohi]

Thank you for the feedback. I will try to improve my habit of saying OK 🙂

[@Omsip123]

@CodeWithAarohi what I also should have written (please apologize), many thanks for your efforts and the good content you provide here on KZbin.

[@soravsingla8782]

Hi Aarohi, your content is excellent and your channel is one of the best Artificial Intelligence channel but still not getting that much of likes which your channel deserves. Hope you succeed #AI #ArtificialIntelligence #DataScience #EducationalContent

[@CodeWithAarohi]

Thank you so much for your kind words and support! It means a lot to me. 😊🙏

[@ZhCrypto]

What does mean by image segmentation and where is code

[@CodeWithAarohi]

Image segmentation means putting a mask on the selected object . Code : github.com/AarohiSingla/Auto-Annotation-Using-YOLOv8-and-SAm