jtydhr88

jtydhr88 / ComfyUI-See-through

Public

A ComfyUI plugin that wraps [See-through](https://github.com/dmMaze/see-through) — an AI system that decomposes a single anime illustration into manipulatable 2.5D layer-decomposed models with depth ordering, ready for Live2D workflows.

60
3
100% credibility
Found Apr 01, 2026 at 59 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Jupyter Notebook
AI Summary

A ComfyUI plugin that decomposes a single anime illustration into up to 24 semantic transparent layers with automatic depth estimation and PSD export for Live2D workflows.

How It Works

1
🎨 Discover See-through for ComfyUI

You find this fun tool that turns a single anime drawing into editable layers like hair, face, and clothes.

2
🛠️ Add it to your ComfyUI

Place the plugin in your ComfyUI folder, install a few helpers, and restart to see new SeeThrough tools appear.

3
🧠 Load the smart models

Click to load the layer and depth models—they download automatically the first time.

4
📤 Upload your anime image

Drag in your favorite character illustration and connect it to the workflow.

5
✨ Watch it decompose magically

Hit queue prompt, and it splits the image into up to 24 transparent layers with smart depth ordering.

6
👀 Preview the blended result

See a perfect reconstruction of your original image from the new layers.

🎉 Download layered PSD

Click the button to get a ready-to-edit PSD file for Live2D or parallax effects.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 59 to 60 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ComfyUI-See-through?

This ComfyUI plugin wraps the See-through AI system to decompose a single anime illustration into up to 24 semantic transparent layers—like hair, face, eyes, clothing—with automatic depth maps for proper ordering. Users load an image, run the decomposition workflow, and export layered PSDs directly from the browser, skipping Photoshop or GIMP manual work. Built around Jupyter Notebook code but runs seamlessly in ComfyUI setups from the comfyui github comfyanonymous repository.

Why is it gaining traction?

Among comfyui best plugins, it stands out by automating 2.5D layer prep for Live2D, with smart left/right splits for eyes and hair via depth clustering, plus previews and Hugging Face model auto-downloads. Easy comfyui plugin install via git clone to custom_nodes and pip, with pre-made workflows for quick comfyui github examples. Pairs well with comfyui github manager or portable versions, even on AMD via comfyui github amd tweaks.

Who should use this?

Anime artists rigging characters for Live2D who hate hours of layer masking in Krita, Photoshop, or GIMP comfyui plugin alternatives. ComfyUI power users building AI pipelines for image decomposition, especially those sharing comfyui github releases or workflows. Plugin developers forking comfyui plugin development in the comfyui plugin directory.

Verdict

Grab it if you're in anime workflows—solid for prototyping despite 46 stars and 1.0% credibility score signaling early maturity. Docs cover comfyui github install and examples well, but expect tweaks for production; test via comfyui github download first.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.