dmmaze/ballonstranslator

深度学习辅助漫画翻译工具, 支持一键机翻和简单的图像/文本编辑 | Yet another computer-aided comic/manga translation tool powered by deeplearning

View on GitHub ↗Jump to charts ↓

Summary Information

Updated 21 minutes ago
Added to GitGenius on September 13th, 2026
Created on April 9th, 2022
Open Issues & Pull Requests: 228 (+0)
GitHub issues: Enabled
Number of forks: 353
Total Stargazers: 5,120 (+0)
Total Subscribers: 24 (+0)

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Issue API getrepoissuespagesummary failed: 429 Rate limit exceeded. Please try again later.

Detailed Description

BallonsTranslator is a computer-aided translation tool for comics and manga powered by deep learning.

The tool addresses the workflow of translating comics by automating the detection, recognition, and translation of text within panels, then reinserting translated text while attempting to preserve the original formatting, colors, outlines, angles, and alignment. It combines optical character recognition, machine translation, and image inpainting to handle the full pipeline in a single pass. The approach supports both manga and Western comics, with layout optimization for English-to-Chinese and Japanese-to-English translations that uses detected speech bubbles as reference for text placement.

The tool suits translators and fan communities working with comics who want to accelerate the initial translation pass, though the README emphasizes that machine-translated results should be clearly marked as such if shared publicly without human review. It includes image editing with masking and inpainting brushes, and text editing with WYSIWYG rich formatting, text effects, and deformation options. The tool also supports context-aware translation through LLM integration, allowing translators to maintain consistency across pages by referencing translation history, terminology tables, and previous completed pages within a configurable token budget. Users can import and export Word documents and perform find-and-replace operations across full text, source, or translated content.

The project maintains active engagement with its user community through public communication channels and accepts pull requests for feature additions. Development includes ongoing refinement of layout handling for specific language pairs, expansion of text styling capabilities, and integration of advanced LLM features for maintaining narrative consistency across longer works.