← Computer Vision learning pathComputer Vision · Module 5 of 10
Object detection
Find and locate multiple objects with bounding boxes.
📖 Key ideas
- Detection outputs class labels plus bounding boxes.
- Popular models: YOLO, SSD, Faster R-CNN, DETR.
- IoU (Intersection over Union) measures box overlap.
- Non-max suppression removes duplicate boxes.
- mAP is the standard detection metric.
💡 Real-life example: Object detection is a teacher circling every mistake on a page and labelling each one.
Bounding boxRectangle around an object
YOLOYou Only Look Once, fast detector
IoUBox overlap measure
mAPMean Average Precision