In plain words: This upgrade tweaks a fast object-spotting network that finds everything in an image in a single pass. At 320x320 it runs in 22 ms, matching a popular rival detector's accuracy at three times the speed.
Abstract
We present some updates to YOLO! We made a bunch of little design changes to make it better. We also trained this new network that's pretty swell. It's a little bigger than last time but more accurate. It's still fast though, don't worry. At 320x320 YOLOv3 runs in 22 ms at 28.2 mAP, as accurate as SSD but three times faster. When we look at the old .5 IOU mAP detection metric YOLOv3 is quite good. It achieves 57.9 mAP@50 in 51 ms on a Titan X, compared to 57.5 mAP@50 in 198 ms by RetinaNet, similar performance but 3.8x faster. As always, all the code is online at https://pjreddie.com/yolo/
Joseph Redmon, Ali Farhadi
arXiv:1804.02767 · cs.CV · submitted Apr 8, 2018
abstract · pdf · html · Tech Report