# senlinuc/caffe_ocr

主流ocr算法研究实验性的项目，目前实现了CNN+BLSTM+CTC架构

Repository: https://github.com/senlinuc/caffe_ocr
Canonical: https://ross.abutalabs.com/products/caffe_ocr
Language: C++
License Family: other
Topics: ocr, lstm, ctc, res-lstm, densenet
Last push: 2020-06-13T13:22:38+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": 3252, "days_push": 2272, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1305, forks 523 (observed 2026-08-28T04:04:18.742814+00:00)

## What it is
An experimental research project built on Caffe implementing CNN+BLSTM+CTC text recognition architectures, with modifications for LSTM, warp-ctc, multi-label data layers, and Inception/ResNet/DenseNet backbones. It includes training pipelines, synthetic Chinese/English data generation, and pretrained OCR models.

## Use cases
- recognize text in images with a CNN+BLSTM+CTC model
- train a Chinese OCR model on synthetic data
- experiment with densenet and resnet OCR architectures in Caffe
- benchmark OCR accuracy and inference speed on CPU and GPU
- adapt Caffe for variable-length sequence recognition with CTC loss

## When to choose
- you want to study or reproduce classic CRNN-style OCR architectures in Caffe
- you need pretrained Chinese/English OCR models with accuracy benchmarks
- you work on Windows with VS2015 and want a ready Caffe OCR setup

## When to avoid
- you need a maintained production OCR system - the project is experimental and inactive since 2020
- you prefer modern frameworks like PyTorch or dedicated OCR tools like Tesseract/PaddleOCR
- you need a permissively licensed project - no license is specified

## Facets
- artifact type: library
- maturity: maintenance
- function: ocr, machine-learning, deep-learning, image-processing
- domain: computer-vision, image-processing, deep-learning
- platform: windows, cpp
- tags: caffe, crnn, blstm, ctc, densenet, resnet, chinese-ocr, text-recognition, research, natural-language-processing, linux, gpu

## Member repositories
- senlinuc/caffe_ocr (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:18.742814+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 2026-08-30T04:51:04.577781+00:00, confidence not recorded.
  - readme: https://github.com/senlinuc/caffe_ocr (fetched 2026-08-28T04:04:18.742814+00:00, sha 87a46f6e4cb5)
- Data as of 2026-08-30T08:39:29.467469+00:00.