January 13, 2021 Software Open Access

ultralytics/yolov3: v9.1 - YOLOv5 Forward Compatibility Updates

Glenn Jocher; Yonghye Kwon; guigarfr; perry0418; Josh Veitch-Michaelis; Ttayu; Daniel Suess; Fatih Baltacı; Gabriel Bianconi; IlyaOvodov; Marc; e96031413; Chang Lee; Dustin Kendall; Falak; Francisco Reveriano; FuLin; GoogleWiki; Jason Nataprawira; Jeremy Hu; LinCoce; LukeAI; NanoCode012; NirZarrabi; Oulbacha Reda; Piotr Skalski; SergioSanchezMontesUAM; Shiwei Song; Thomas Havlik; Timothy M. Shead

This release is a minor update implementing numerous bug fixes, feature additions and performance improvements from https://github.com/ultralytics/yolov5 to this repo. Models remain unchanged from v9.0 release.

Branch Notice

The ultralytics/yolov3 repository is now divided into two branches:

• Master branch: Forward-compatible with all YOLOv5 models and methods (recommended).

  $ git clone https://github.com/ultralytics/yolov3  # master branch (default)
  

• Archive branch: Backwards-compatible with original darknet *.cfg models (⚠️ no longer maintained).

  $ git clone -b archive https://github.com/ultralytics/yolov3  # archive branch
  

<img src="https://user-images.githubusercontent.com/26833433/100382066-c8bc5200-301a-11eb-907b-799a0301595e.png" width="800">

** GPU Speed measures end-to-end time per image averaged over 5000 COCO val2017 images using a V100 GPU with batch size 32, and includes image preprocessing, PyTorch FP16 inference, postprocessing and NMS. EfficientDet data from google/automl at batch size 8.

Pretrained Checkpoints Model AP^val AP^test AP₅₀ Speed_GPU FPS_GPU params FLOPS YOLOv3 43.3 43.3 63.0 4.8ms 208 61.9M 156.4B YOLOv3-SPP 44.3 44.3 64.6 4.9ms 204 63.0M 157.0B YOLOv3-tiny 17.6 34.9 34.9 1.7ms 588 8.9M 13.3B

AP^test denotes COCO test-dev2017 server results, all other AP results denote val2017 accuracy. All AP numbers are for single-model single-scale without ensemble or TTA. Reproduce mAP by python test.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65
Speed_GPU averaged over 5000 COCO val2017 images using a GCP n1-standard-16 V100 instance, and includes image preprocessing, FP16 inference, postprocessing and NMS. NMS is 1-2ms/img. Reproduce speed by python test.py --data coco.yaml --img 640 --conf 0.25 --iou 0.45 All checkpoints are trained to 300 epochs with default settings and hyperparameters (no autoaugmentation). Test Time Augmentation (TTA) runs at 3 image sizes. Reproduce TTA** by python test.py --data coco.yaml --img 832 --iou 0.65 --augment

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt