Insightface: Please list the Glint360K's face alignment settings

Created on 23 Oct 2020 · 1Comment · Source: deepinsight/insightface

can we download the original unaligned face images ? thanks !

Source

apache2046

👍1

Most helpful comment

Alignment settings is:

import cv2
from skimage import transform as trans
dst = np.array([
   [30.2946, 51.6963],
   [65.5318, 51.5014],
   [48.0252, 71.7366],
   [33.5493, 92.3655],
   [62.7299, 92.2041]], dtype=np.float32 )
dst[:,0] += 8.0
src = landmark.astype(np.float32)
tform = trans.SimilarityTransform()
tform.estimate(src, dst)
M = tform.params[0:2, :]
img = cv2.warpAffine(img, M, (112,112), borderValue=0.0)

This is same as all datasets in insightface.

anxiangsir on 23 Oct 2020

👍8 🎉5

>All comments

Alignment settings is:

import cv2
from skimage import transform as trans
dst = np.array([
   [30.2946, 51.6963],
   [65.5318, 51.5014],
   [48.0252, 71.7366],
   [33.5493, 92.3655],
   [62.7299, 92.2041]], dtype=np.float32 )
dst[:,0] += 8.0
src = landmark.astype(np.float32)
tform = trans.SimilarityTransform()
tform.estimate(src, dst)
M = tform.params[0:2, :]
img = cv2.warpAffine(img, M, (112,112), borderValue=0.0)

This is same as all datasets in insightface.

anxiangsir on 23 Oct 2020

👍8 🎉5

Was this page helpful?

0 / 5 - 0 ratings