-
After landmark is obtained, how to get aligned faces suitable for face recognition?
-
Hi, I am wondering if there is extracted the face and align it part inside your code?
If not, how did you do this part?
-
MoltenVK SDK version is 1.3.290
Full Shader here:
```glsl
#version 460
#extension GL_EXT_buffer_reference : require
layout(std430, buffer_reference, buffer_reference_align = 4) readonly buff…
-
I got face locations from face.bbox but how can i display it?
"""
top, right, bottom, left = face.bbox.astype(np.int)
face_image = img[top:bottom, left:right]
pil_image = Image.fromarray(fa…
-
The diorama camera setup should take into account the avatar face/eyes metrics, to ensure the avatar head fits. Right now it's tuned for normal human-sized avatars, but we can pull the full metrics fr…
-
![image](https://user-images.githubusercontent.com/10508053/98925702-4e1b0f00-2511-11eb-9cb3-b8a539177328.png)
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Problem description
When using the Alight to selection function, the camera rotates at a strange angle
### Fu…
-
I found that in predict.py only the croped face are put into net to predict the attributes, the face is not aligned, is it right?
-
as the title
-
Hi @davidsandberg ,
I'm studying the Machine Learning on Face Recognition recently and I found that your project is a great start so I'm reading all the python codes recently.
However, I found …