📃: Segment anything on images.

[jaidh01]

🔴 Title: Segment anything in real time
🔴 Aim: to develop a model, where we can segment anything in real time.
🔴 Brief Explanation:

We will use Fastsam for segmentation, because of its faster responses, in this we would be able to segment anything on an image.
The user will get 4 options: -segment everything, segment by box-prompt, point-prompt and text-prompt.

Screenshots 📷

---

✅ To be Mentioned while taking the issue :

• Full name : Jai Dhingra
• What is your participant role? GSSOC, Hacktoberfest

---

Happy Contributing 🚀

All the best. Enjoy your open source journey ahead. 😎

[Avdhesh-Varshney]

Don't raise more than 1 issue

[jaidh01]

@Avdhesh-Varshney, I have raised a PR for my previous issue. So, I request you to please assign this issue to me.

[Avdhesh-Varshney]

Can you explain the keyword segment according to your context? Properly elaborate your project

[jaidh01]

@Avdhesh-Varshney
In the context of my project Segment Anything on Images, the term segment refers to a computer vision task where an image is divided into meaningful regions or segments. The goal of segmentation is to identify and separate objects or areas of interest within the image. Each pixel in the image is classified as belonging to a specific segment or object. This differs from object detection, which merely places bounding boxes around objects; segmentation goes deeper by identifying the precise boundaries and shapes of objects.

Use Case: This can be particularly useful in applications like medical imaging, autonomous driving, and augmented reality, where understanding the exact shape and boundaries of objects is crucial.

[jaidh01]

Although, now I figured out a way to implement it in real time as well.

[Avdhesh-Varshney]

Based on your use case, I'm assigned it to you