# KUR-creative/SickZil-Machine

Manga/Comics Translation Helper Tool

Repository: https://github.com/KUR-creative/SickZil-Machine
Canonical: https://ross.abutalabs.com/products/sickzil-machine
Language: Python
License: AGPL-3.0
License Family: copyleft
Topics: manga, comics, deep-learning, image-processing
Last push: 2022-11-22T04:11:38+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2754, "days_push": 1380, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1524, forks 162 (observed 2026-08-28T04:04:58.160370+00:00)

## What it is
SickZil-Machine is a desktop application that automates text removal from manga and comic pages during the scanlation (translation) process. It uses two deep learning models — a U-net-based segmentation network to detect text regions and a Deepfill v2-based completion network to erase text and reconstruct the background naturally.

## Use cases
- remove text from manga pages automatically
- clean comic panels for scanlation typesetting
- erase speech bubbles text while preserving background art
- prepare manga images for translation typesetting
- batch inpaint text regions in comic scans

## When to choose
- you are scanlating manga or comics and need to erase original text
- you want fully automatic text detection and removal without manual editing
- you need natural background reconstruction after text removal

## When to avoid
- you need text translation or typesetting itself — this only removes text
- you need a polished, actively maintained tool — it is a pre-release with ongoing groundwork
- you work with non-comic images where manga-trained models may underperform

## Facets
- artifact type: application
- maturity: experimental
- function: image-processing, machine-learning, deep-learning, ocr
- domain: image-processing, computer-vision, deep-learning, media
- platform: python, cross-platform
- tags: manga, scanlation, comics, text-removal, image-inpainting, translation-helper, desktop

## Member repositories
- KUR-creative/SickZil-Machine (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:58.160370+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 2026-08-30T04:31:41.605585+00:00, confidence not recorded.
  - readme: https://github.com/KUR-creative/SickZil-Machine (fetched 2026-08-28T04:04:58.160370+00:00, sha 617359917f4b)
- Data as of 2026-08-30T08:39:29.467469+00:00.