> For the complete documentation index, see [llms.txt](https://captic-2.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://captic-2.gitbook.io/docs/overview/good-to-know/ai-vision-explained.md).

# AI-Vision Explained

Before we dive in, let’s make sure you know *why* you're choosing AI-powered vision. 

## AI-Vision vs. Traditional Vision

**Traditional Vision** systems follow strict rules — you have to *program* them to handle every variation. And let’s be honest, sometimes those rules get *out of hand*. Trying to account for *all* the edge cases? Not fun. Not feasible.

Enter **AI-Vision**. Instead of programming rules, you *teach* the system. If it’s seen the variations before in the dataset, it can handle them. No extra coding needed.

#### In Short:

* **Traditional Vision** is programmed
* **AI-Vision** learns.

## When to use AI-Vision

If your traditional system is throwing in the towel, AI-Vision is here to save the day.

## What AI-Vision Can Do

AI-powered vision systems offer a lot of flexibility and power. Some core capabilities include:

* **Classification**: Sort things into categories.
* **Object Detection**: Find and identify objects in your scene.
* **Segmentation**: Pinpoint the exact shape and boundaries of objects.
* **Optical Character Recognition:** Read text in an image.

<figure><img src="https://482575286-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F22HCD2cZG8ud6cPRBzVQ%2Fuploads%2FgCBREj7rLgej5WWlKxdB%2Fimage.png?alt=media&amp;token=5a90a4d9-c92c-4b11-9988-0d3859002d2c" alt=""><figcaption></figcaption></figure>

AI-Vision can also be combined with more traditional techniques to make them, like:

* **Dimension Calculation**
* **Barcode Detection**
* **...**