How to build an Image Similarity Search app with Image Embeddings & Qdrant

Показать описание

In this video, I'll show you how to use ResNet's Image Model to convert a dataset of images into a series of embeddings (or vectors!), that we can then upload to a vector database - we'll be using Qdrant Cloud. From there, we can then query our embeddings using our database; we can even search for similar records!

What we'll cover
===

🔎 Sourcing an image dataset (We'll be using Kaggle to fetch ours)
🌆 Image Embeddings (We'll use Microsoft's ResNet 50 Model)
📊 Vector Databases (We'll use Qdrant Cloud to host our data!)
💻 Streamlit (for the frontend of our app)

Timestamps
===

0:00 Introduction
0:28 What are we building?
0:58 How will we build it?
3:01 Converting our images to embeddings
16:49 Uploading our embeddings to the vector database
22:30 Building the frontend with Streamlit
35:59 Outtro