all projects
COCO · 90% mAP

YOLOv3 Object Detection

A real-time object-detection pipeline built with YOLOv3 and non-max suppression on the COCO dataset.

Result ~90% mAP on COCO Tags YOLOv3 · NMS · COCO · OpenCV
View on GitHub

A real-time object-detection pipeline implementing YOLOv3 from the ground up.

Highlights

  • Single-shot detection with the YOLOv3 architecture.
  • Non-max suppression to resolve overlapping boxes into clean detections.
  • Evaluated on COCO, reaching roughly 90% mAP.
  • Built to understand the detection stack end-to-end — anchors, grid cells, and the confidence/NMS post-processing — rather than calling a black-box detector.