Akihito Miyazaki's picture
2

Akihito Miyazaki PRO

Akjava

AI & ML interests

I'm developing a user-friendly, browser-based platform that allows users to connect various AI services like ChatGPT, Gemini, local LLMs, Hugging Face models, and more. Our goal is to empower users to build custom AI tools by seamlessly combining these services, similar to Langchain or ComfyUI.

Recent Activity

Articles

Organizations

None yet

Akjava's activity

reacted to davidberenstein1957's post with 🔥 28 days ago
view post
Post
1701
Let’s make a generation of amazing image-generation models

The best image generation models are trained on human preference datasets, where annotators have selected the best image from a choice of two. Unfortunately, many of these datasets are closed source so the community cannot train open models on them. Let’s change that!

The community can contribute image preferences for an open-source dataset that could be used for building AI models that convert text to image, like the flux or stable diffusion families. The dataset will be open source so everyone can use it to train models that we can all use.

Blog: https://huggingface.co/blog/burtenshaw/image-preferences
posted an update about 1 month ago
posted an update about 1 month ago
view post
Post
491
Finaly I realesed mediapipe-face animation space.

Mediapipe 68-points Eyes-Closed and Mouth-Opened
Akjava/mediapipe-68-facial-guide-eyes-closed-mouth-opened

[Article]Results: Converted Guide Images(eyes-closed and mouth-opened) with Flux.1 schenll img2img/inpaint
https://huggingface.co/blog/Akjava/result-guide-image-eyes-mouth

All the other tools listed are designed to support Mediapipe Face Animation

Akjava/mediapipe-tools-672ffe8ee7b62763c31b70c7

Akjava/webp-3-frame-talking-animation-tools-672819ce4989f354cdbcc739
posted an update about 1 month ago
view post
Post
554
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
posted an update about 2 months ago
view post
Post
701
I've released several new Hugging Face Spaces.

My primary objective is to create consistent character facial animation using image-to-image techniques:

Akjava/CreateConsistentCharacterFacialAnimationWithImg2Img

A short-term goal is create simple talk-head animation.

WebP-3-Frame-Talking-Animation
Akjava/AIDiagramChatWithVoice-FaceCharacter

[Space]

- GPU tools
Flux1-schnell img2img
Akjava/flux1-schnell-img2img

Flux1-schnell Inpaint with mask-file
Akjava/flux1-schnell-img2img

- Tiny CPU tools
WebP-3F-TH - create webp animation from 3 images
OpenCV-Inapint - classic inpaint
Whitebalance - simple white balance
Paste Image - just paste image with mask
WebP Resize Convert - resize and convert webp-animation
posted an update 3 months ago
view post
Post
1419
Streaming Text-to-Speech Chat Demo (CPU Inference Client)

Akjava/mistral-7b-v0.3-matcha-tts-en

Please be patient, as it may take over a minute to load the ONNX model.

This demo utilizes an inference client, which may occasionally become unresponsive.