dmMaze/BallonsTranslator
BallonsTranslator
BallonsTranslator is a software suite designed for extracting, translating, and replacing text within comic panels while preserving the original visual layout. It functions as an image translation tool that combines text region detection, optical character recognition, and deep learning inpainting to automate the localization of comics.
The tool features a deep learning image inpainter that removes original text and restores backgrounds using generative neural networks and patch-matching algorithms. It also includes a rich-text translation editor for modifying translated dialogue with support for font presets, search-and-replace, and document exports.
The system provides a multi-engine OCR pipeline for extracting text and font colors, and a layout-aware replacement system that matches font sizes and positioning. For automated workflows, a headless command-line interface allows for batch image translation and rendering without a graphical user interface.
Features
- Comic Panel Translators - Detects, removes, and replaces text in comic panels using deep learning while preserving the original visual layout.
- Image Inpainting - Ships a deep learning inpainting pipeline that removes text and restores comic backgrounds.
- Image Text Translators - Extracts, translates, and replaces text in comic panels while preserving the original visual layout using deep learning inpainting.
- Batch Processing - Automates the full translation pipeline across multiple images via a headless command-line interface.
- Translation APIs - Provides a multi-engine translation abstraction that switches between cloud APIs and local models.
- OCR Pipelines - Extracts text and font colors from comic panels using multiple recognition engines in an end-to-end pipeline.
- OCR Engines - Extracts text from detected regions using multiple OCR engines with optional font name detection.
- Text Detection Models - Uses specialized deep learning models to detect text regions in comic panels for Japanese and English.
- Headless Batch Pipelines - Runs the full translation pipeline in headless batch mode via a configuration file.
- Layout-Aware Replacement - Matches font sizes and positioning to insert translated dialogue while preserving original comic layout.
- Headless Batch Pipelines - Runs the full translation pipeline in headless batch mode via a configuration file.
- Comic Text Extraction - Identifies text areas in comic panels using specialized models for Japanese and English.
- Translation Editors - Includes a rich-text translation editor with font presets, search-and-replace, and document exports.
- Translation Text Editors - Adjusts font formatting and style presets, and performs find-and-replace across source or translated text.
