# shitagaki-lab/see-through

"Single-image Layer Decomposition for Anime Characters" (SIGGRAPH 2026 Conference Paper)

Repository: https://github.com/shitagaki-lab/see-through
Canonical: https://ross.abutalabs.com/products/see-through
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-05T13:31:48+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 96, release rhythm 35, longevity 11
- inputs: {"age_days": 155, "days_push": 28, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3641, forks 332 (observed 2026-08-28T04:08:12.815199+00:00)

## What it is
A research framework from a SIGGRAPH 2026 paper that decomposes a single anime character illustration into up to 23 fully inpainted, semantically distinct layers with inferred drawing order, enabling 2.5D manipulation. It is a Python/PyTorch codebase with optional annotator tiers and online demos.

## Use cases
- decompose an anime illustration into editable layers
- turn a static anime image into a 2.5D model
- extract hair, face, eyes, and clothing layers from one image
- generate inpainted PSD layers from anime art
- reorder and animate anime character parts from a single picture

## When to choose
- you need layer separation for anime-style artwork specifically
- you want a research-grade, open-source implementation with a paper backing
- you have a CUDA or ROCm GPU for inference

## When to avoid
- you need layer decomposition of photorealistic images rather than anime characters
- you want a polished end-user GUI application rather than a Python research codebase
- you have no GPU and cannot run heavy deep-learning inference locally

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning, computer-vision, graphics
- domain: computer-vision, image-processing, graphics, machine-learning
- platform: python, windows
- tags: layer-decomposition, anime, 2-5d-models, image-inpainting, siggraph, research-code, psd, gpu, linux, macos

## Member repositories
- shitagaki-lab/see-through (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:12.815199+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:31:47.516746+00:00, confidence not recorded.
  - readme: https://github.com/shitagaki-lab/see-through (fetched 2026-08-28T04:08:12.815199+00:00, sha 5848be09c2f7)
- Data as of 2026-08-30T08:39:29.467469+00:00.