Ultra Light
Ultra Light Fast Generic Face Detector 👨👩👧👦🖼
Very fast and quality face detector. Can use CPU, GPU and MPS (Apple M1 ML) providers. Work via ONNX model
Installation
pip install ultralight
Usage sample
import cv2
from ultralight import UltraLightDetector
from ultralight.utils import draw_faces
image = cv2.imread('sample.jpg')
detector = UltraLightDetector()
boxes, scores = detector.detect_one(image)
print(f'Found {len(boxes)} face(s)')
# >>> Found 14 face(s)
draw_faces(image, boxes, scores)
cv2.imshow('result', image)
cv2.waitKey(0)
This sample can be found here
Reference
GitHub repository of original detector
ArXiv paper of original detector
More
PyPI: https://pypi.org/project/ultralight
Repository: https://github.com/abionics/UltraLight
Developer: Alex Ermolaev (Abionics)
Email: abionics.dev@gmail.com
License: MIT (see LICENSE.txt)
1 maintainer
Metadata
1 maintainer
Release files for ultralight 2.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| ultralight-2.2.0.tar.gz | 8.0 kB | Details |
Release files / ultralight-2.2.0.tar.gz
| Download URL | ultralight-2.2.0.tar.gz |
|---|---|
| Size | 8.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f6e5e0e7c638bb6ea76a0f87dba20e792484aba0667a8f78a80effcacee1c67f
|
|
BLAKE2b-256 checksum How to use checksums |
5a0b02612654e972a90d733df209badabb7ad38af8718bb2e6cdc6e12fd53d50
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.10.6
|
