Hands-on computer vision projects — open-vocabulary detection, video tracking, and domain-specific fine-tuning, built from real models and real failure cases.
Each numbered folder is a self-contained project: code, notebook, and its own README.
| # | Project | What it covers |
|---|---|---|
| 01 | Open-Vocabulary Detection | Why fixed-vocabulary detectors (YOLO) mislabel unfamiliar objects, and three ways to fix it |
More projects in progress — video object tracking, and an end-to-end chess vision pipeline.
Each project has its own README.md with specific setup instructions.