Progress Checklist¶
Tick off each chapter after you've done its exercises, and each capstone once you can complete it without notes.
These checkboxes don't save
The boxes are visual only on the website. To track your progress, copy this page into your own notes or repository.
🌱 Beginner tier: Images and Classical Vision¶
Level 0: Images and tools¶
- What is computer vision?
- Setting up your vision toolkit
- Digital images
- Images as arrays
- Working with images and video
- Capstone: Build a photo-processing tool: batch-convert, resize, annotate, and analyze a folder of images
Level 1: Image processing¶
- Point operations and histograms
- Convolution and filtering
- Edges and gradients
- The frequency domain
- Morphology and classical segmentation
- Geometric transformations
- Capstone: Build a document scanner: find the page, rectify it, and clean it up
Level 2: Classical computer vision¶
- Features and keypoints
- Matching and RANSAC
- Cameras and calibration
- Stereo and multi-view geometry
- Motion and tracking
- Classical recognition
-
Capstone: Stitch a panorama and measure depth: features, RANSAC, homographies, and stereo
-
Tier checkpoint passed (see the tier page)
🔧 Intermediate tier: Deep Learning for Vision¶
Level 3: CNNs for vision¶
- From pixels to learned features
- Convolutional networks in depth
- Training image classifiers
- Transfer learning and fine-tuning
- Understanding and debugging CNNs
- Capstone: Train an image classifier on a new domain, and explain what it learned
Level 4: Object detection¶
- Detection fundamentals
- Two-stage detectors
- One-stage detectors
- Transformers for detection
- Detection in practice
- Capstone: Build and evaluate an object detector end to end
Level 5: Segmentation and dense prediction¶
- Semantic segmentation
- Instance and panoptic segmentation
- Keypoints and pose estimation
- Depth and other dense tasks
- Promptable segmentation
-
Capstone: Train a segmentation model and measure it properly
-
Tier checkpoint passed (see the tier page)
🚀 Expert tier: Modern and 3D Vision¶
Level 6: Transformers and multimodal vision¶
- Vision transformers
- Self-supervised learning for vision
- Vision-language models
- Open-vocabulary and foundation models
- Capstone: Build a zero-shot image search and classification system with CLIP-style embeddings
Level 7: Generative vision¶
- Image generation foundations
- Diffusion models for images
- Controllable generation and editing
- Capstone: Train a small diffusion model that generates images, and steer it with guidance
Level 8: Video, 3D, and deployment¶
- Video understanding
- Multi-object tracking
- 3D vision and point clouds
- Neural rendering
- Deploying vision models
-
Capstone: Ship a real-time video analytics pipeline: detect, track, count, and optimize
-
Tier checkpoint passed (see the tier page)