See-through: Anime Character Layer Decomposition
Skill by ara.so — Daily 2026 Skills collection.
See-through is a research framework (SIGGRAPH 2026, conditionally accepted) that decomposes a single anime illustration into up to 23 fully inpainted, semantically distinct layers with inferred drawing orders — exporting a layered PSD file suitable for 2.5D animation workflows.
What It Does
- Decomposes a single anime image into semantic layers (hair, face, eyes, clothing, accessories, etc.)
- Inpaints occluded regions so each layer is complete
- Infers pseudo-depth ordering using a fine-tuned Marigold model
- Exports layered
.psdfiles with depth maps and segmentation masks - Supports depth-based and left-right stratification for further refinement
Installation
Optional Annotator Tiers
Install only what you need:
Always run all scripts from the repository root as the working directory.
Models
Models are hosted on HuggingFace and downloaded automatically on first use:
Key CLI Commands
Main Pipeline: Layer Decomposition to PSD
Output is saved to workspace/layerdiff_output/ by default. Each run produces:
- A layered
.psdfile with semantically separated layers - Intermediate depth maps
- Segmentation masks
Heuristic Post-Processing
After the main pipeline, further split layers using heuristic_partseg.py:
Synthetic Training Data Generation
Python API Usage
Running the Full Pipeline Programmatically
Batch Processing a Directory
Post-Processing: Depth and LR Splits
Loading and Inspecting PSD Output
Interactive Body Part Segmentation (Notebook)
Open and run the provided demo notebook:
This demonstrates interactive 19-part body segmentation with visualization using the SAM body parsing model.
Dataset Preparation for Training
See-through uses Live2D model files as training data. Setup requires a separate repo:
Launching the UI
Directory Structure
Common Patterns
Pattern: End-to-End Single Image Workflow
Pattern: Check Available Body Tags
ComfyUI Integration
A community-maintained ComfyUI node is available:
Install via ComfyUI Manager or clone into ComfyUI/custom_nodes/.

