Image Classification Using Vision Transformer | An Image is Worth 16x16 Words

ExplainingAI · Advanced ·👁️ Computer Vision ·2y ago

About this lesson

This video covers the implementation of vision Transformer - VIT in pytorch . This is the third part of Vision transformer - VIT series in which I build entire vision transformer from scratch. I also cover visualizations of the attention map using attention rollout as well as positional embedding visualization to get more intuition on what the model is learning in this video. I have covered Vision Transformer VIT in three parts: 1. Patch Embedding Module - https://youtu.be/lBicvB4iyYU 2. Attention Block - https://www.youtube.com/watch?v=zT_el_cjiJw 3. Building Vision Transformer and visualizing attention map using attention rollout and also visualizing positional embedding of vision transformer (This Video) *Other Good Resources* Yannic Kilcher | An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (Paper Explained) - https://www.youtube.com/watch?v=TrdevFK_am4 AI Coffee Break with Letitia | An image is worth 16x16 words: ViT | Vision Transformer explained - https://www.youtube.com/watch?v=DVoHvmww2lQ James Briggs | Vision Transformers (ViT) Explained + Fine-tuning in Python - https://www.youtube.com/watch?v=qU7wO02urYU Good Place to understand general transformer further - https://tinyurl.com/exai-vit-transformer Timestamps : 00:00 Intro 00:28 Architecture of Single Vision Transformer Layer 01:19 Architecture of Vision Transformer 02:00 Dataset for Image Classification 02:35 Code for Vision Transformer Layer 03:48 Code & Implementation of VIT - Vision Transformer 04:50 Attention Map Visualization using Attention Rollout 06:09 Attention Maps for trained model 06:18 Position Embedding Visualization 08:32 Outro Paper Link - https://tinyurl.com/exai-vit-paper Implementation - https://tinyurl.com/exai-vit-code Subscribe to Channel - https://tinyurl.com/exai-channel-link Good Place to understand original transformer further - https://tinyurl.com/exai-vit-transformer Background Track - Fruits of Life by Jimena Contreras Email - explainingai.

Original Description

This video covers the implementation of vision Transformer - VIT in pytorch . This is the third part of Vision transformer - VIT series in which I build entire vision transformer from scratch. I also cover visualizations of the attention map using attention rollout as well as positional embedding visualization to get more intuition on what the model is learning in this video. I have covered Vision Transformer VIT in three parts: 1. Patch Embedding Module - https://youtu.be/lBicvB4iyYU 2. Attention Block - https://www.youtube.com/watch?v=zT_el_cjiJw 3. Building Vision Transformer and visualizing attention map using attention rollout and also visualizing positional embedding of vision transformer (This Video) *Other Good Resources* Yannic Kilcher | An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (Paper Explained) - https://www.youtube.com/watch?v=TrdevFK_am4 AI Coffee Break with Letitia | An image is worth 16x16 words: ViT | Vision Transformer explained - https://www.youtube.com/watch?v=DVoHvmww2lQ James Briggs | Vision Transformers (ViT) Explained + Fine-tuning in Python - https://www.youtube.com/watch?v=qU7wO02urYU Good Place to understand general transformer further - https://tinyurl.com/exai-vit-transformer Timestamps : 00:00 Intro 00:28 Architecture of Single Vision Transformer Layer 01:19 Architecture of Vision Transformer 02:00 Dataset for Image Classification 02:35 Code for Vision Transformer Layer 03:48 Code & Implementation of VIT - Vision Transformer 04:50 Attention Map Visualization using Attention Rollout 06:09 Attention Maps for trained model 06:18 Position Embedding Visualization 08:32 Outro Paper Link - https://tinyurl.com/exai-vit-paper Implementation - https://tinyurl.com/exai-vit-code Subscribe to Channel - https://tinyurl.com/exai-channel-link Good Place to understand original transformer further - https://tinyurl.com/exai-vit-transformer Background Track - Fruits of Life by Jimena Contreras Email - explainingai.
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related Reads

📰
How to Self-Host Nextcloud: A Complete Setup Guide
Learn to self-host Nextcloud for private file storage and synchronization, and understand why it matters for data privacy
Dev.to · stack_versus
📰
Is OneOpen Annotator a Practical Free Alternative to Roboflow?
Discover if OneOpen Annotator is a viable free alternative to Roboflow for computer-vision workflows and learn where each option shines
Medium · Machine Learning
📰
How Vision AI Is Transforming Industrial Operations with Existing Cameras
Learn how Vision AI transforms industrial operations using existing cameras, increasing efficiency and accuracy
Medium · Machine Learning
📰
Agentic vision: Building visual intelligence with Amazon Bedrock and MCP servers
Learn how to build visual intelligence with Amazon Bedrock and MCP servers for streamlined AI capabilities
AWS Machine Learning

Chapters (10)

Intro
0:28 Architecture of Single Vision Transformer Layer
1:19 Architecture of Vision Transformer
2:00 Dataset for Image Classification
2:35 Code for Vision Transformer Layer
3:48 Code & Implementation of VIT - Vision Transformer
4:50 Attention Map Visualization using Attention Rollout
6:09 Attention Maps for trained model
6:18 Position Embedding Visualization
8:32 Outro
Up next
9-Phase Computer Vision Roadmap 2026 | AI & Deep Learning | #shorts
SCALER
Watch →