Playing with Azure AI Vision

This guide offers a comprehensive introduction to Azure AI Vision, providing users with step-by-step instructions to effectively leverage its capabilities.

Hafeez Baig
31 steps
3 minutes

Azure

This lab is a continuation of the previous one, Please refer to it before proceeding - <https://scribehow.com/viewer/Exploring_Azure_AI_Foundry__Creating_Hub_and_Project__VIf2tVybTYCeZ6AOuinV4g>

Make sure you are on Azure AI Foundry page, click on the resource project which you created in the previous lab

**Overview** page will open, here you can view the Endpoints, keys and project details

Click on **Model catalog** from the left bar, here you can view all the AI models

Click on **Playgrounds** from the left bar

Scroll the page and click on the **Try the** **Language playground** button

**What is Language Playground in Azure AI Foundry?** The Language Playground in Azure AI Foundry is an interactive environment that enables users to experiment with large language models (LLMs) like GPT-4o-mini. It allows you to test prompts, configure system messages, and observe model responses in real-time, facilitating rapid prototyping and refinement of AI applications. This tool is particularly useful for developers aiming to fine-tune model behavior before deploying applications to production environments.

Here you can view and configure language playground

Click on **AI Services** from the left bar

Scroll the page and click on the **Vision + Document** option

Scroll the page to the **view all other capabilities** section and click on the **Image** tab

Click on the **Try demo** option for **Common object detection**

**What is Common object detection in Azure AI Services?** Common Object Detection in Azure AI Services refers to the capability of identifying and locating frequently seen objects, such as people, vehicles, or furniture, in images or videos. This is enabled through Azure’s Computer Vision or Custom Vision services. The service returns object labels along with bounding boxes, allowing applications to understand visual content for tasks like surveillance, retail analytics, or automation.

**Detect common objects in images** page will open, click on any sample image

Here you can view the **Detected attributes** on the right side about the sample image

Now go back to the **Vision + Document** page

Click on the **Try demo** option for **Image captioning**

**What is Image captioning in Azure AI Services?** Image captioning in Azure AI Services is the process of generating descriptive text for an image using AI models. Azure's Computer Vision service analyzes the visual content and produces human-like captions that summarize the scene. This is useful for accessibility, content indexing, and enhancing search experiences.

**Add captions to image** page will open, click on any sample image

On the right side you can view the **Detected attributes** about the sample image

Click on the **JSON** tab to view the JSON format

Go back to the **Vision + Document** page, click on the **Try demo** option for **Dense captioning**

**What is Dense captioning in Azure AI Services?** Dense captioning in Azure AI Services refers to generating multiple detailed descriptions for different regions within an image. Unlike standard image captioning, which provides a single overall description, dense captioning breaks the image into segments and describes each part individually. This enables deeper image understanding, useful in applications like surveillance, content moderation, and scene analysis.

**Dense captioning** page will open, click on any sample image

On the right side you can view the **Detected attributes** about the sample image

Click on the **JSON** tab to view the JSON format

Scroll the page to the **Run the code** section and click on the **View Code** button

Here you can view the sample code

Click on the Python dropdown on the top right side of the wizard, here you can select C# language

On the same page, click on the **View documentation** page

Documentation will be opened in a new tab, here you can refer the documentation **Note:** Each service has their own documentation.

Go back to the **Azure AI** browser tab

Click on the **Try demo** option for **Optical character recognition**

**What is Optical character recognition in Azure AI Services?** Optical Character Recognition (OCR) in Azure AI Services is a feature that extracts printed or handwritten text from images, PDFs, or scanned documents. It enables automated reading of text for tasks like digitizing receipts, invoices, ID cards, or forms, making content searchable and usable in applications such as data entry, compliance, and document processing.

**Extract text from images** page will open, click on any sample image

On the right side you can view the **Detected attributes** about the sample image

Click on the **JSON** tab to view the JSON format

Congratulations on completing this lab!

Refer to this lab to continue learning - <https://scribehow.com/viewer/Exploring_Azure_AI_Document_Intelligence__Demo__Fw1IJoULS2ShRlLBl3esBw>

Playing with Azure AI Vision

This guide offers a comprehensive introduction to Azure AI Vision, providing users with step-by-step instructions to effectively leverage its capabilities.

Hafeez Baig
31 steps
3 minutes

Azure

This lab is a continuation of the previous one, Please refer to it before proceeding - <https://scribehow.com/viewer/Exploring_Azure_AI_Foundry__Creating_Hub_and_Project__VIf2tVybTYCeZ6AOuinV4g>

Make sure you are on Azure AI Foundry page, click on the resource project which you created in the previous lab

**Overview** page will open, here you can view the Endpoints, keys and project details

Click on **Model catalog** from the left bar, here you can view all the AI models

Click on **Playgrounds** from the left bar

Scroll the page and click on the **Try the** **Language playground** button

Here you can view and configure language playground

Click on **AI Services** from the left bar

Scroll the page and click on the **Vision + Document** option

Scroll the page to the **view all other capabilities** section and click on the **Image** tab

Click on the **Try demo** option for **Common object detection**

**Detect common objects in images** page will open, click on any sample image

Here you can view the **Detected attributes** on the right side about the sample image

Now go back to the **Vision + Document** page

Click on the **Try demo** option for **Image captioning**

**Add captions to image** page will open, click on any sample image

On the right side you can view the **Detected attributes** about the sample image

Click on the **JSON** tab to view the JSON format

Go back to the **Vision + Document** page, click on the **Try demo** option for **Dense captioning**

**Dense captioning** page will open, click on any sample image

On the right side you can view the **Detected attributes** about the sample image

Click on the **JSON** tab to view the JSON format

Scroll the page to the **Run the code** section and click on the **View Code** button

Here you can view the sample code

Click on the Python dropdown on the top right side of the wizard, here you can select C# language

On the same page, click on the **View documentation** page

Documentation will be opened in a new tab, here you can refer the documentation **Note:** Each service has their own documentation.

Go back to the **Azure AI** browser tab

Click on the **Try demo** option for **Optical character recognition**

**Extract text from images** page will open, click on any sample image

On the right side you can view the **Detected attributes** about the sample image

Click on the **JSON** tab to view the JSON format

Congratulations on completing this lab!

Refer to this lab to continue learning - <https://scribehow.com/viewer/Exploring_Azure_AI_Document_Intelligence__Demo__Fw1IJoULS2ShRlLBl3esBw>