Algorithm python library linked to vsource platform.
Project description
链接VSOURCE_FACE_PLATFORM ,采用平台里的RESTFUL API提供一套本地能够使用的算法库。
pip install vsource -i https://pypi.python.org/simple
目前功能:
人脸识别:输入两张人脸图像判断来自一个人的概率
说话人识别:输入两个音频判断来自一个人的概率
人脸检测:输入一张图像检测出所有的人脸框图
人脸属性:输入一张图像检测人脸并判断人脸的表情和年龄
一个人脸识别的Demo:
import vsource
if __name__ == '__main__':
username = {{ secrets.username }}
password = {{ secrets.password }}
vsource.login(username, password)
face_path1 = 'examples/0006_01.jpg'
face_path2 = 'examples/0007_01.jpg'
score = vsource.face_recognition(face_path1, face_path2)
print(score)
一个说话人识别的Demo:
import vsource
if __name__ == '__main__':
username = {{ serects.username }}
password = {{ serects.password }}
vsource.login(username, password)
audio_path1 = 'examples/0.wav'
audio_path2 = 'examples/1.wav'
score = vsource.speaker_recognition(audio_path1, audio_path2)
print(score)
一个人脸检测的Demo:
import vsource
if __name__ == '__main__':
username = {{ serects.username }}
password = {{ serects.password }}
vsource.login(username, password)
face_path1 = 'examples/0008_01.jpg'
result = vsource.face_detection(face_path1)
print(result)
一个人脸属性的Demo:
import vsource
if __name__ == '__main__':
username = {{ serects.username }}
password = {{ serects.password }}
vsource.login(username, password)
face_path1 = 'examples/0008_01.jpg'
result = vsource.face_attribute(face_path1)
print(result)
# 其他同学实现的版本
result2 = vsource.face_attribute(face_path1, version='fsx')
print(result)
TIPS:
关于用户名和密码,防止恶意的请求进入,导致服务器收到大量的请求后排队时间过长进一步让服务都不可用,所以暂时还是需要登录态,关于试用的用户名和密码可以联系我。
持续的更新各种算法中。
算法如果遇到超时,可以设置参数max_interval=x秒,每个算法都带这个参数,比如face_recognition(face_path, max_interval=100)。如果长时间没有结果,说明算法运行时出现了错误。
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
vsource-1.1.7.tar.gz
(15.9 kB
view details)
File details
Details for the file vsource-1.1.7.tar.gz
.
File metadata
- Download URL: vsource-1.1.7.tar.gz
- Upload date:
- Size: 15.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.4.1 importlib_metadata/4.5.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.13
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 878f8f4b88ebb67db62467d2f6cac888b275b8a5449bca46bb4e8e10ebbb606f |
|
MD5 | 0c928e8df564968c7d7fef6d052d0568 |
|
BLAKE2b-256 | 73b6a4d7f806d9105bdff87f0bfe12c11a42ac3fecaa788560666939bf00a284 |