OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Sickzil Machine

Sickzil Machine is a Windows desktop application that uses deep-learning models to find text in manga and comic images and erase it by reconstructing the surrounding background. It supports automatic cleanup and lets users specify an area to remove.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Sickzil Machine — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it as a text-removal and background-reconstruction step when preparing manga or comic images for translation. The documented workflow does not include typesetting.

Who it is for

  • Manga and comic translators preparing pages for translated lettering.
  • Scanlation editors who need to clear original text before later editing or typesetting.

Documented uses

  • Detect text regions in comic pages.
  • Remove lettering and fill the cleared area to match the background.
  • Specify a particular area for erasure when needed.
Before you choose

Requirements & limits

The repository describes a Windows release and links to a 0.1.1 pre-release. It is licensed under AGPL-3.0.

  • The application combines SegNet text-region detection with ComplNet background completion; the README associates these with U-net and Deepfill v2 approaches.
  • The README describes both detection and removal as automatic, with an option to specify an erase area.
  • Developer instructions list Python, TensorFlow 1.13 requirements, NVIDIA driver 410.x, CUDA 10.0 and cuDNN 7.4.1 or later.
  • The README says the program is not perfect and invites bug reports and suggestions.
  • Typesetting is identified as future work rather than a documented current feature.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up SickZil-Machine for developer execution

Prepare the repository and its release resources, then start the documented Python entry point.

Before you start

  • NVIDIA driver 410.x, CUDA 10.0, and cuDNN 7.4.1 or later
  • A release ZIP downloaded from the SickZil-Machine releases page

Steps

  1. Clone the repository and change into its directory.

  2. Download a release ZIP, unzip it, and copy its resource/cnet and resource/snet directories into the repository's resource directory.

  3. Install the Python dependencies with the command below.

    Documented command
    pip install -r requirements.txt

  4. Change into src and run the command below.

    Documented command
    python main.py

Expected outcome

The documented developer entry point is launched; the README does not specify further execution details or a verification procedure.

Guide limitations

  • The listed NVIDIA, CUDA, and cuDNN prerequisites are specified for TensorFlow 1.13.0.
  • SickZil-Machine is described as not perfect, and the README lists improving text-segmentation performance as future work.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
KUR-creative/SickZil-Machine · Official homepage title · retained in its source language
How we review sourcesSuggest a correction