README.md

YOLOv3: An Incremental Improvement

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.

Citation

@misc{redmon2018yolov3,
    title={YOLOv3: An Incremental Improvement},
    author={Joseph Redmon and Ali Farhadi},
    year={2018},
    eprint={1804.02767},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Results and Models

Backbone	Scale	Lr schd	Mem (GB)	Inf time (fps)	box AP	Config	Download
DarkNet-53	320	273e	2.7	63.9	27.9	config	model | log
DarkNet-53	416	273e	3.8	61.2	30.9	config	model | log
DarkNet-53	608	273e	7.4	48.1	33.7	config	model | log

Mixed Precision Training

We also train YOLOv3 with mixed precision training.

Backbone	Scale	Lr schd	Mem (GB)	Inf time (fps)	box AP	Config	Download
DarkNet-53	608	273e	4.7	48.1	33.8	config	model | log

Lightweight models

Backbone	Scale	Lr schd	Mem (GB)	Inf time (fps)	box AP	Config	Download
MobileNetV2	416	300e	5.3		23.9	config	model | log
MobileNetV2	320	300e	3.2		22.2	config	model | log

Notice: We reduce the number of channels to 96 in both head and neck. It can reduce the flops and parameters, which makes these models more suitable for edge devices.

Credit

This implementation originates from the project of Haoyu Wu(@wuhy08) at Western Digital.