"""
Overview:
Detect persons in both real photo and anime images.
Trained with `deepghs/anime_person_detection <https://huggingface.co/datasets/deepghs/anime_person_detection>`_ \
and open-sourced real photos datasets.
.. image:: person_detect_demo.plot.py.svg
:align: center
This is an overall benchmark of all the person detect models:
.. image:: person_detect_benchmark.plot.py.svg
:align: center
The models are hosted on
`huggingface - deepghs/real_person_detection <https://huggingface.co/deepghs/real_person_detection>`_.
"""
from typing import List, Tuple
from imgutils.data import ImageTyping
from imgutils.generic import yolo_predict
_REPO_ID = 'deepghs/real_person_detection'
[docs]def detect_persons(image: ImageTyping, model_name: str = 'person_detect_v0_s_yv11',
conf_threshold: float = 0.35, iou_threshold: float = 0.7, **kwargs) \
-> List[Tuple[Tuple[int, int, int, int], str, float]]:
"""
Detect persons in both real photo and anime images using YOLO models.
This function applies a pre-trained YOLO model to detect persons in the given anime image.
It supports different model levels and versions, allowing users to balance between
detection speed and accuracy.
:param image: The input image for person detection. Can be various image types supported by ImageTyping.
:type image: ImageTyping
:param model_name: Optional custom model name. If provided, it overrides the auto-generated model name.
:type model_name: str
:param conf_threshold: The confidence threshold for detections. Only detections with confidence
scores above this threshold will be returned. Default is 0.35.
:type conf_threshold: float
:param iou_threshold: The Intersection over Union (IoU) threshold for non-maximum suppression.
Detections with IoU above this threshold will be merged. Default is 0.7.
:type iou_threshold: float
:return: A list of detected persons. Each person is represented by a tuple containing:
- Bounding box coordinates as (x0, y0, x1, y1)
- The string 'person' (as this function only detects persons)
- The confidence score of the detection
:rtype: List[Tuple[Tuple[int, int, int, int], str, float]]
:example:
>>> from realutils.detect import detect_persons
>>>
>>> detect_persons('yolo/solo.jpg')
[((0, 30, 398, 599), 'person', 0.926707923412323)]
>>> detect_persons('yolo/2girls.jpg')
[((0, 74, 760, 1598), 'person', 0.7578195333480835), ((437, 33, 1200, 1600), 'person', 0.6875205039978027)]
>>> detect_persons('yolo/3+cosplay.jpg')
[((106, 69, 347, 591), 'person', 0.8794167041778564), ((326, 14, 592, 534), 'person', 0.8018194437026978), ((167, 195, 676, 675), 'person', 0.5351650714874268)]
>>> detect_persons('yolo/multiple.jpg')
[((1305, 441, 1891, 1534), 'person', 0.8789498805999756), ((206, 191, 932, 1533), 'person', 0.8423126935958862), ((1054, 170, 1417, 1055), 'person', 0.8138357996940613), ((697, 659, 1473, 1534), 'person', 0.7926754951477051), ((685, 247, 1128, 1526), 'person', 0.5261526703834534), ((690, 251, 1125, 1126), 'person', 0.4193646311759949)]
>>> from imgutils.detect import detection_visualize
>>> from matplotlib import pyplot as plt
>>>
>>> image = 'yolo/solo.jpg'
>>> result = detect_persons(image)
>>>
>>> # visualize it
>>> plt.imshow(detection_visualize(image, result))
>>> plt.show()
"""
return yolo_predict(
image=image,
repo_id=_REPO_ID,
model_name=model_name,
conf_threshold=conf_threshold,
iou_threshold=iou_threshold,
**kwargs,
)