How to Build Text-Based Image Search with LLMs | Powered by Microsoft Foundry

Description

✨ Learn how to build Text-Based Image Search using LLM powered Image Verbalization and models deployed in Microsoft Foundry. What you’ll discover: 🔥 📌 How vector spaces, similarity, and data modalities actually work 📌 The difference between image verbalization and multimodal embedding models 📌 How to use LLM + embedding models deployed in Microsoft Foundry 📌 How to authenticate securely using RBAC + DefaultAzureCredential 📌 How to verbalize images and embed descriptions in C# 📌 How to run Text-Based image search 📝 Blog post: https://deployedinazure.com/image-verbalization-via-llm/ ----- 0:00 Intro 1:32 What We're Building Today 4:10 Vectors, Vector Spaces, Data Modalities 11:35 Image Verbalization vs. Multimodal Embeddings 16:06 LLM and Embedding Model in Microsoft Foundry 19:19 Secure Access to Microsoft Foundry using RBAC 24:43 Verbalizing Images using LLM + Embeddings (C# Example) 31:01 Analyzing the Results 34:42 Summary ----- #microsoftfoundry #azuredeveloper #azureai Follow me 👉 🌐 Blog: https://deployedinazure.com 💻 GitHub: https://github.com/deployed-in-azure 🔗 LinkedIn: https://www.linkedin.com/company/deployed-in-azure