-
이미지 pre-processing:
- MTCNN로 bounding box와 keypoint 추출
- MTCNN으로 여러개의 bounding box가 발견됐을 때 처리 (넓이가 가장 큰 것만 쓰기)
- 눈과 코 위치를 사용하여 왼쪽 눈 오른쪽 눈이 수평을 이루도록 rotate (예: skimage 패키지의 similarity transform 사용…
-
## Summary
Add new task for facial recognition to identify individuals.
## Datasets
**NOTE:** recommendations here are not vetted
- [DigiFace-1M](https://paperswithcode.com/dataset/digiface-1m)
D…
-
dear anson0910:
I am so appreciated your work. I have a question about the 3000 images. Is there a trick to select these negative images? for example: 1. Keep various 2. use the residue part of…
-
### Feature description
The current indexing method cannot guarantee that each image will correctly replace the intended face when replacing faces in group photos in bulk. Could you add a face detect…
-
Thanks for sharing the code!
I tried the checkpoint [IN256_Base](https://huggingface.co/TencentARC/Open-MAGVIT2/blob/main/imagenet_256_B.ckpt) on face images, but the faces in the reconstructed ima…
-
I am trying to adapt this library to support the face swapping of still images but having some problems getting it working. I've gotten the face scanning to work with user uploaded still images. Does…
-
@BeauNouvelle How can we use this magic in SwiftUI's `Image` view?
-
Description: The default bounding box for face detection correctly covers the face on iPad, but the landmarks are positioned outside the face and appear rotated by 90 or 270 degrees.
Steps to Repro…
-
hello, when I run
```bash download.sh ./laion_face_meta ./laion_face_data```
tha download fails
```
total - success: 0.005 - failed to download: 0.995 - failed to resize: 0.000 - images p…
-
* {{ cookiecutter.project_name }} version: test_cookiecutter_maryam_5
* Python version: remote machine 2.7.13
* Operating System: openSUSE Leap
### Description
trying to run the docker creatio…