Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Akjava 
posted an update Nov 15, 2024
Post
565
hi All I just shared Spaces and Article.

This key feature is Mediapipe face landmarker
Apache Licensed and trained with own dataset.
Good licensed model who use Flux.1 schnell instead of Dev

[Spaces]
Mediapipe Face detect
Akjava/mediapipe-face-detect

Face crop and replace
Akjava/mediapipe-face-crop-and-replace

Mediapipe 68 landmark
Akjava/mediapipe-68-points-facial-landmark

Mediapipe 68 Face Mask
Akjava/mediapipe-68-points-facial-mask

[Articles]
Better img2img results with Flux.1 schnell by using ScaleUp or Sharpen or FillColor pre-processing
https://huggingface.co/blog/Akjava/img2img-pre-processing
In this post
@Akjava on Hugging Face: "hi All I just shared Spaces and Article. This key feature is Mediapipe face…"

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Akjava 
posted an update Nov 15, 2024
Post
565
hi All I just shared Spaces and Article.

This key feature is Mediapipe face landmarker
Apache Licensed and trained with own dataset.
Good licensed model who use Flux.1 schnell instead of Dev

[Spaces]
Mediapipe Face detect
Akjava/mediapipe-face-detect

Face crop and replace
Akjava/mediapipe-face-crop-and-replace

Mediapipe 68 landmark
Akjava/mediapipe-68-points-facial-landmark

Mediapipe 68 Face Mask
Akjava/mediapipe-68-points-facial-mask

[Articles]
Better img2img results with Flux.1 schnell by using ScaleUp or Sharpen or FillColor pre-processing
https://huggingface.co/blog/Akjava/img2img-pre-processing
In this post