How to Train a One-Class Detector from 300 Photos

Most tutorials skip the hard parts. This episode walks through the full process of training a single-class object detector from scratch: how many images you actually need, why scene-based splitting matters more than random splits, how to annotate consistently, and why YOLOv8n is the right architecture for small datasets. Using the concrete example of detecting a specific anti-graffiti poster design city-wide, we cover the metrics that matter when every missed detection costs $200 — and the honest signal for when to give up and use a general model instead. Episode #242417 — open it directly at myweirdprompts.com/242417