pyFaceTrace 5.0.3

easy Face Recognition for python

Author:KuoYuan Li

本程式簡單地結合dlib,opencv
讓不懂機器學習的朋友可以軟簡單地操作人臉辨識,
程式需另外安裝 dlib
搜尋關鍵字：whl dlib cp*** ***代表python版本 ex:cp310代表 python 3.10
dlib whl 安裝包下載網站: (https://github.com/datamagic2020/Install-dlib)

• 本套工具主要針對windows使用者設計，相依之 package 及相容性問題需自行排除
• dlib whl 安裝包下載後必需由檔案離線安裝 pip install ...
• opencv whl 下載點:請下載合適的opencv版本
  (https://pypi.tuna.tsinghua.edu.cn/simple/opencv-contrib-python/)

※PS:
2022/11/24 使用 python3.10 搭配
dlib-19.22.99-cp310-cp310-win_amd64.whl 試用成功

2023/4/2
調整函式，適配 colab
(https://colab.research.google.com/drive/1ou7nWLQGl8uYLR_jUDyush9-D8ToTe8P?usp=sharing)
移除不常用之影像檔處理函式
新增直接和 opencv 協作模式

Download the samples to 'train' folder(下載各種照片樣本至train資料夾)

import pyFaceTrace as ft  
ft.downloadImageSamples()  

work with opencv webcam process (detail)

import pyFaceTrace as ft
import cv2
from PIL import ImageFont
ft.loadDB()
cap = cv2.VideoCapture(0,cv2.CAP_DSHOW)
while True:
    ret, img = cap.read()
    FONT = ImageFont.truetype("kaiu.ttf",50,index=0)
    rect=ft.detector(img,1)[0]
    fv=ft.getFeatureVector(img,rect)
    tag,dist = ft.predictFromDB(fv)
    cv2.rectangle(img,(rect.left(),rect.top()),(rect.right(),rect.bottom()),(0,0,255),3)
    img=ft.addText2Img_cv2(img,tag,FONT,position=(rect.left(),rect.top()-FONT.size-1))
    if cv2.waitKey(10) == 27: break
    cv2.imshow('press esc to exit...', img)
cap.release()
cv2.destroyAllWindows()

work with opencv webcam process (esay)

import pyFaceTrace as ft
import cv2
ft.loadDB()
cap = cv2.VideoCapture(0,cv2.CAP_DSHOW)
while(True):    
    ret, img = cap.read()
    if not ret:continue
    #img=cv2.flip(img,1)
    tags,dists,rects,img = ft.predictImage(img)            
    cv2.imshow('press esc to exit...', img)
    if cv2.waitKey(10) == 27: break
    
cap.release()
cv2.destroyAllWindows()

Demo with webcam

import pyFaceTrace as ft
ft.loadDB(folder='train')
ft.predictCam()

比對目前webcam擷取到的人臉和指定影像檔案並計算它們之間的距離

import pyFaceTrace as ft  
im = ft.captureImageFromCam()
VTest = ft.getFeatureVector(im)
Vtrain = ft.loadFeatureFromPic('train\\李國源.jpg')
D=ft.dist(VTest,Vtrain)
print('距離=',D)

載入train資料夾中所有jpg檔之特徵及tag並直接預測目前webcam擷取到的人臉對應的TAG

import pyFaceTrace as ft
ft.loadDB(folder='train')
im = ft.captureImageFromCam()
VTest = ft.getFeatureVector(im)
result = ft.predictFromDB(VTest)
print(result)

License

MIT