A minimal PyTorch implementation of YOLOv3, with support for inference. This is a wrapper of YOLOV3-pytorch implementation here as a standalone python package.

$ pip install git+https://github.com/mkocabas/yolov3-pytorch.git

Uses pretrained weights to make predictions on images.

$ python examples/detect.py --image_folder data/samples/

Below table displays the inference times when using as inputs images scaled to 256x256.

from yolov3.yolo import YOLOv3
from torchvision.transforms import Compose, ToTensor, Normalize

detector = YOLOv3(device='cuda', img_size=608)
preprocess_img = Compose([
        ToTensor(),
        Normalize(mean=[0.485, 0.456, 0.406],
                  std=[0.229, 0.224, 0.225])
])

img = preprocess_img('data/sample.png').to('cuda')

predictions = detector(img)

• YOLOv3: An Incremental Improvement

Joseph Redmon, Ali Farhadi

[Paper] [Project Webpage] [Authors' Implementation]

@article{yolov3,
  title={YOLOv3: An Incremental Improvement},
  author={Redmon, Joseph and Farhadi, Ali},
  journal = {arXiv},
  year={2018}
}

• YOLOv3 PyTorch implementation: link