YOLO9000: Better, Faster, Stronger

CVPR 2017 Joseph RedmonAli Farhadi

We introduce YOLO9000, a state-of-the-art, real-time object detection system that can detect over 9000 object categories. First we propose various improvements to the YOLO detection method, both novel and drawn from prior work... (read more)

PDF Abstract


Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Object Detection COCO YOLO v2 + Darknet-19 Bounding Box AP 21.6 # 44
Object Detection PASCAL VOC 2007 YOLO v2 MAP 78.6% # 9