# Sanster/text_renderer

Generate text images for training deep learning ocr model

Repository: https://github.com/Sanster/text_renderer
Canonical: https://ross.abutalabs.com/products/text_renderer
Language: Python
License: MIT
License Family: permissive
Topics: synthtext, crnn, ocr
Last push: 2022-01-17T07:58:32+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3019, "days_push": 1689, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1464, forks 383 (observed 2026-08-28T04:04:47.861831+00:00)

## What it is
A Python tool that generates synthetic text images with configurable visual effects for training deep learning OCR models like CRNN. It supports both Latin and non-Latin scripts, with strict mode to ensure fonts cover all corpus characters.

## Use cases
- generate synthetic training data for an OCR model
- create text images with perspective and blur effects for CRNN training
- render Chinese text images with fonts that support all characters
- augment OCR datasets with varied text effects
- produce labeled text image datasets for deep learning

## When to choose
- you need labeled synthetic text images to train an OCR recognizer
- you work with non-Latin scripts and need font coverage guarantees
- you want configurable effects like curves, borders, and blur on rendered text

## When to avoid
- you need a pretrained OCR model rather than training data
- you require actively maintained software - the project points to a newer fork (oh-my-ocr/text_renderer)
- you need scene-text detection datasets rather than rendered text

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, data-generation, machine-learning, ocr
- domain: computer-vision, machine-learning, image-processing, deep-learning
- platform: python, cli
- tags: synthetic-data, ocr-training, text-rendering, crnn, data-augmentation, linux

## Member repositories
- Sanster/text_renderer (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:47.861831+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 2026-08-30T04:35:08.678734+00:00, confidence not recorded.
  - readme: https://github.com/Sanster/text_renderer (fetched 2026-08-28T04:04:47.861831+00:00, sha 977652230e7d)
- Data as of 2026-08-30T08:39:29.467469+00:00.