Skip to content
Projects

andandandand/practical-computer-vision

View on GitHub

andandandand/practical-computer-vision has 204 GitHub stars and 66 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-17 15:33 UTC. Last push 2026-09-18.

How these counts are measured

Measured star growth

in 24h
0
in 7 days
Collecting history
in 15 days
Collecting history
in 1 month
Collecting history

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

About

Practical Computer Vision Bootcamp with notebooks, quizzes, and videos

Machine Learning

README.md

Original

Hands-on Computer Vision Bootcamp

Visualizing image embeddings in a 2D space through the FiftyOne app

This repository provides tutorials for people at various stages of learning computer vision. It covers foundational concepts, such as image representation and the basics of neural networks, and extends to more advanced topics including generative models and zero-shot learning. The material is designed to be useful for those new to the field as well as practitioners looking to work with contemporary tools and techniques.

🛠️ Get Started: Setting Up Your Environment

Before starting the notebooks, prepare your environment using these setup guides:

📚 Tutorial Notebooks

This collection of tutorial notebooks is organized to progress from foundational concepts to advanced applications. Click the badges to open them in Google Colab or Kaggle.

Module / Topic Notebook Title Description Run Online
1. Foundations Digital Image Representation Understand how images are represented in PIL, NumPy, and PyTorch. Open In Colab Kaggle
2. Introduction to FiftyOne Getting Started with FiftyOne Datasets Learn to load, explore, and visualize image datasets using FiftyOne. Open In Colab
3. Basic Neural Networks MLP for Image-based Regression Train a Multilayer Perceptron (MLP) for regression tasks based on image features. Open In Colab Kaggle
Matrix Multiplication, Activations, & Network Shape Explore neural network mechanics: matrix math, non-linearities, and network architecture. Open In Colab Kaggle
Predicting Car Prices From Images Apply MLP regression to predict car prices from their images. Open In Colab Kaggle
4. Convolutional Neural Networks (CNNs) Fundamentals of CNNs with LeNet5 Implement LeNet5 for digit recognition (Kaggle competition). Open In Colab Kaggle
Looking into LeNet5 with Random Weights Investigate LeNet5 behavior with random weight initialization. Open In Colab Kaggle
Dropout Visualization on MNIST Understand and visualize dropout regularization on MNIST images. Open In Colab Kaggle
5. Advanced Classification & Transfer Learning Classifying Pet Images by Breed Tackle image classification: identifying pet breeds. Open In Colab Kaggle
Pretrained ResNets for Classification Use pre-trained ResNet models (ImageNet) for image labeling. Open In Colab Kaggle
Fine-tuning a ResNet for Multilabel Classification Adapt a pre-trained ResNet for multilabel image tasks. Open In Colab Kaggle
Interpretability with Class Activation Mapping (CAM) See which image parts contribute to CNN classification. Open In Colab Kaggle
6. Image Embeddings & Similarity Intuitions about Image Embeddings Understand image embeddings and pairwise comparisons. Open In Colab Kaggle
Extracting Embeddings with Pretrained ResNet34 Generate image embeddings using a pre-trained ResNet34. Open In Colab Kaggle
Visualizing Embeddings & Clustering with FiftyOne Use FiftyOne to visualize and cluster image embeddings. Open In Colab
Visualizing Image Embeddings with TensorBoard Use TensorBoard for visualizing high-dimensional embedding spaces. Open In Colab Kaggle
Image Neighborhoods & Clustering with sklearn Find similar images and cluster street artwork using scikit-learn. Open In Colab Kaggle
7. Image Segmentation Semantic Segmentation with U-Net Perform pixel-level classification (semantic segmentation) using U-Net. Open In Colab Kaggle
8. Zero-Shot Learning Intro to CLIP & Zero-Shot Classification Discover CLIP for zero-shot image classification. Open In Colab Kaggle
Zero-Shot Object Detection with Grounding DINO Perform object detection for new classes using Grounding DINO. Open In Colab Kaggle
9. Generative Models Denoising Diffusion U-Net for MNIST Generation Explore DDPMs with a U-Net for generating MNIST digits. Open In Colab
Artwork Generation with Stable Diffusion XL Generate images using the Stable Diffusion XL model. Open In Colab
10. Utilities & Data Collection Scraping Public Domain Artwork from Bing Images Scrape images from Bing and save to Google Drive for custom datasets. Open In Colab

🎓 Full Course with Evaluations

For a structured learning experience with quizzes and assignments, enroll in the full course on openHPI:

🎬 Watch the Series on YouTube

Video tutorials are available:

💬 Join the Community

For questions, tutorial discussions, or project sharing, join the KINETO.AI Slack workspace. The channel for this content is #practical-computer-vision-bootcamp.

Contributing, Contact & License

You are welcome to fork this repository, experiment with the code, and contribute. Please maintain the Creative Commons Attribution License.

Author: Antonio Rueda-Toicen

antonio 'at' kineto 'dot' ai

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Happy learning!