ku21fan/STR-Fewer-Labels
Scene Text Recognition (STR) methods trained with fewer real labels (CVPR 2021)
This project helps computer vision practitioners accurately extract text from challenging real-world images, such as photos of signs or labels. It takes an image containing scene text as input and outputs the recognized text. This tool is designed for machine learning engineers and researchers who develop and deploy scene text recognition systems.
184 stars. No commits in the last 6 months.
Use this if you need to train highly accurate scene text recognition models using significantly less real-world labeled data than traditional methods, achieving performance comparable to or better than models trained on vast synthetic datasets.
Not ideal if you are looking for a pre-trained, ready-to-use API for simple text extraction without needing to train or fine-tune models yourself.
Stars
184
Forks
29
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Dec 23, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ku21fan/STR-Fewer-Labels"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
JaidedAI/EasyOCR
Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin,...
breezedeus/CnSTD
CnSTD: 基于 PyTorch/MXNet 的 中文/英文 场景文字检测（Scene Text Detection）、数学公式检测（Mathematical Formula...
githubharald/SimpleHTR
Handwritten Text Recognition (HTR) system implemented with TensorFlow.
felixdittrich92/OnnxTR
OnnxTR a docTR (Document Text Recognition) library Onnx pipeline wrapper - for seamless,...
mindee/doctr
docTR (Document Text Recognition) - a seamless, high-performing & accessible library for...