elliottzheng / face-detection Goto Github PK

@elliottzheng 你好。请问项目里有人脸对齐功能吗？

More time during batch inference as compared to insightface's single pass!!

`
import numpy as np
from insightface.app import FaceAnalysis
import cv2
import matplotlib.pyplot as plt
import time
from batch_face import RetinaFace

img = cv2.imread('/home/zeeshan/Downloads/avengers.png')
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
img = cv2.resize(img, (640, 640))
plt.imshow(img)

model = FaceAnalysis(allowed_modules=['detection'],
providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
model.prepare(ctx_id=0)
tik = time.time()
faces = model.get(img)
faces = model.get(img)
faces = model.get(img)
faces = model.get(img)
faces = model.get(img)
print(f"time taken: {time.time()-tik}")

detector = RetinaFace(gpu_id=0)
tik = time.time()
faces = detector.detect([img, img, img, img, img])
print(f"time taken: {time.time()-tik}")
`

I have even tried repeating the same img object inside the batch array and then computing the result. Even in that case the total of computing individually with insightface is much less than this? @elliottzheng could you please have a look at this?

Hello!

Great work BTW, I really like the implementation. I am trying to use your package in one of my projects.

Would it be possible for you to upload the face_detection package to PyPI to improve useability and deployment ease?

I can also upload it under your name if you provide me permission to do so.

在detector = RetinaFace(gpu_id=0)中，gpu_id可以是多个值吗，我试着让gpu_id=[0,1,2,3]但遇到错误

After a fresh install using pip, the model weights file seems to be missing from the installation. My guess is that the MANIFEST.in needs to include files ending with .pth for pip install to work.

Hi!
According to the changelog, I found that this code had a Resnet version before.
Is it still available? Thanks!