Learn How to build your custom Image Segmentation model using YOLO-NAS and Segment Anything Model (SAM).
For Queries: You can comment in comment section or you can mail me at aarohisingla1987@gmail.com
YOLO-NAS is a new real-time state-of-the-art object detection model that outperforms both YOLOv6 & YOLOv8 models in terms of mAP (mean average precision) and inference latency.
The “NAS” stands for “Neural Architecture Search,” a technique used to automate the design process of neural network architectures. Instead of relying on manual design and human intuition, NAS employs optimization algorithms to discover the most suitable architecture for a given task.
Github:
SuperGradients GitHub:
Starter Notebook:
Training with SuperGradients, Deci's open-source, PyTorch-based computer vision library, enables advanced techniques like Distributed Data Parallel, Exponential Moving Average, Automatic mixed precision, and Quantization Aware Training.
SuperGradients is fully compatible with PyTorch Datasets and Dataloaders, so you can use your dataloaders as is.
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.
We can try a demo on their site :
Github code is also available : …
You can download the dataset also: …
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
#objectdetection #imagesegmentation #yolo-nas #computervision #sam #segmentanythingmodel
