Nessuna descrizione

113 Commit

Niels c835652b37 Add HF		10 mesi fa
assets	6158471016 ini	10 mesi fa
docker	1c68863560 upload docker	10 mesi fa
examples	1c0ef1216a examples	10 mesi fa
logs	7e52e5f450 logs	10 mesi fa
tests	c4a0a465fc test	10 mesi fa
ultralytics	c835652b37 Add HF	10 mesi fa
LICENSE	2758ec06b2 Initial commit	10 mesi fa
README.md	db33aa16c5 Update README.md	10 mesi fa
app.py	db51919405 demo	10 mesi fa
mkdocs.yml	dd89c57437 mkdocs	10 mesi fa
pyproject.toml	4adbe10cf2 upload pyproject	10 mesi fa
requirements.txt	ed2b434655 env	10 mesi fa

YOLOv12

YOLOv12: Attention-Centric Real-Time Object Detectors

[Yunjie Tian](https://sunsmarterjie.github.io/)¹, [Qixiang Ye](https://people.ucas.ac.cn/~qxye?language=en)², [David Doermann](https://cse.buffalo.edu/~doermann/)¹ ¹ University at Buffalo, SUNY, ² University of Chinese Academy of Sciences.

Comparison with popular methods in terms of latency-accuracy (left) and FLOPs-accuracy (right) trade-offs

Updates

2025/02/20: Any computer or edge device?. Support yolov12 now.
2025/02/20: ONNX CPP Version. Train a yolov12 model on a custom dataset?
2025/02/19: arXiv version is public. Demo is available.

Abstract

Enhancing the network architecture of the YOLO framework has been crucial for a long time but has focused on CNN-based improvements despite the proven superiority of attention mechanisms in modeling capabilities. This is because attention-based models cannot match the speed of CNN-based models. This paper proposes an attention-centric YOLO framework, namely YOLOv12, that matches the speed of previous CNN-based ones while harnessing the performance benefits of attention mechanisms.

YOLOv12 surpasses all popular real-time object detectors in accuracy with competitive speed. For example, YOLOv12-N achieves 40.6% mAP with an inference latency of 1.64 ms on a T4 GPU, outperforming advanced YOLOv10-N / YOLOv11-N by 2.1%/1.2% mAP with a comparable speed. This advantage extends to other model scales. YOLOv12 also surpasses end-to-end real-time detectors that improve DETR, such as RT-DETR / RT-DETRv2: YOLOv12-S beats RT-DETR-R18 / RT-DETRv2-R18 while running 42% faster, using only 36% of the computation and 45% of the parameters.

Main Results

Installation

wget https://github.com/Dao-AILab/flash-attention/releases/download/v2.7.3/flash_attn-2.7.3+cu11torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl
conda create -n yolov12 python=3.11
conda activate yolov12
pip install -r requirements.txt
pip install -e .

Validation

yolov12n yolov12s yolov12m yolov12l yolov12x

from ultralytics import YOLO

model = YOLO('yolov12{n/s/m/l/x}.pt')
model.val(data='coco.yaml', save_json=True)

Training

from ultralytics import YOLO

model = YOLO('yolov12n.yaml')

# Train the model
results = model.train(
  data='coco.yaml',
  epochs=600, 
  batch=256, 
  imgsz=640,
  scale=0.5,  # S:0.9; M:0.9; L:0.9; X:0.9
  mosaic=1.0,
  mixup=0.0,  # S:0.05; M:0.15; L:0.15; X:0.2
  copy_paste=0.1,  # S:0.15; M:0.4; L:0.5; X:0.6
  device="0,1,2,3",
)

# Evaluate model performance on the validation set
metrics = model.val()

# Perform object detection on an image
results = model("path/to/image.jpg")
results[0].show()

Prediction

from ultralytics import YOLO

model = YOLO('yolov12{n/s/m/l/x}.pt')
model.predict()

Export

from ultralytics import YOLO

model = YOLO('yolov12{n/s/m/l/x}.pt')
model.export(format="engine", half=True)  # or format="onnx"

Demo

python app.py
# Please visit http://127.0.0.1:7860

Acknowledgement

The code is based on ultralytics. Thanks for their excellent work!

Citation

@article{tian2025yolov12,
  title={YOLOv12: Attention-Centric Real-Time Object Detectors},
  author={Tian, Yunjie and Ye, Qixiang and Doermann, David},
  journal={arXiv preprint arXiv:2502.12524},
  year={2025}
}

Model	size ^(pixels)	mAP^val 50-95	Speed ^{T4 TensorRT10}	params ^(M)	FLOPs ^(G)
YOLO12n	640	40.6	1.64	2.6	6.5
YOLO12s	640	48.0	2.61	9.3	21.4
YOLO12m	640	52.5	4.86	20.2	67.5
YOLO12l	640	53.7	6.77	26.4	88.9
YOLO12x	640	55.2	11.79	59.1	199.0

README.md