Segment Anything Model (SAM): Build Custom Image Segmentation Model Using YOLOv8 and SAM

Learn How to build your custom Image Segmentation model using Segment Anything Model (SAM).

Segment Anything Model (SAM): a new AI model from Meta AI that can "cut out" any object, in any image, with a single click.

SAM is a promptable segmentation system with zero-shot generalization to unfamiliar objects and images, without the need for additional training.

Meta the parent company of Social Media giant Facebook has launched a Image Segmentation model yesterday known as the "Segment Anything" Model, capable of identifying and extracting individual objects within an image or video.

This model can help us perform segmentation task very easily.
They built the largest segmentation dataset with over 1 billion masks on 11M images.
Meta said they evaluated the model’s capabilities on various tasks and find that its zero-shot performance is impressive (test model on new and unseen scenarios without additional training.) even if it hasn't been specifically trained to recognize them.
Simply we can say: SAM can identify objects that were not a part of its training.

There are different kind of tasks which you can do with this new Segment Anything Model:
You can cut-out the objects from Images.
You can put masks on Objects

[soravsingla]

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

[riebbkk]

Thank you so much for your tutorial.
I would like to request that you make a video on Road lane line detection and road lane detection using any deep-learning model (Segmentation anything model (SAM) + Any deep-learning model like ResNet 50 etc..)
Thank you

[deepikasharmar]

👏👏👏👏👏
Mam waiting for ur video on tensorflow in jetson nano !!!

[mohammedelmehdimehaya]

you amazing i'm one of your followers from long time

[soravsingla]

Thank you so much for sharing this valuable knowledge

[fatemadalal]

Thank you for the tutorial. Found it helpful. Also found a couple of other useful tutorials on your channel will be checking that out as well. I would like to ask if you have experience in reinforcement learning as well and have implemented that specifically for deep learning related tasks such as model training etc. If yes and if you could make video on that it will be helpful. Thank you!!

[wydrrne]

great explanation, very useful output, but is there any possibility to change the segmented wheel of car with another wheel and should be fit only the presented mask area not to overlap with car??

[soravsingla]

Best YouTube channel / Platform to learn Artificial Intelligence, Data Science, Data Analysis, Machine Learning.
#BestChannel #YoutubeChannel #ArtificialIntelligence #CodeWithAarohi #DataScience #Engineering #MachineLearning #DataAnalysis #BestLearning #LearnDataScience #DataScienceCourse #AytificialIntelligenceCourse #Codewithaarohi #CodeWithAarohi Code with Aarohi

[mahejabeenhossain]

That was really helpful! I would love to know how you save the segmentation so it can be used for training later. I am planning on using this to automate my image annotation process.

[dbhwqhx]

Thank you so much for your tutorial！I want to know how can I create my own data to train SAM?

[angelospapadopoulos]

amazing work as always!
We would like to see an implementation of SAM with text prompt (till now we have point and box as input prompt)

[sarahnawoya]

thank you madam for the nice video. how can I extract that mask to use it further in the code? for example to binarize it

[LaviSrivastava-uy]

please provide the notebook file of this video on custom dataset with SAM

[rqswqkp]

Hiii Arohi thanks for the video it helped a lot. I want to know how can we get label on masked and bounding boxed object.

[vivekpaswan]

Getting AttributeError: 'Boxes' object has no attribute 'tolist'.

[ayanbahukhandi]

Hello, Im working on a similar project but instead of truck i want to segment human faces out of an image, i've got to the point where i pass the face to the model, but the model is segmenting different parts of face, like nose, eyes etc etc. its not segmenting the face as a whole. How can i fix it? so that it segments out the whole face.

[vanditaagarwal]

how to get your code? The code which you have showed in this video

[alocatnaf]

thanks for this nice tutorial!
where can we find this jupyter notebook?
thanks alot in advance!

[shinwarikhan]

hello mam .. thank you for such an amazing contents.. we learn much more from your videos... mam can you make video on yolo-NAS and what is yolo-NAS and how it work and how can we detect object using it ..we will be thankful

[nitishrathore]

Ultralytics isn't commerically viable to use, is there any other open source distribution that is free to use commercially and is fast?