argman/EAST

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3059 stars · 1039 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A TensorFlow re-implementation of the EAST (Efficient and Accurate Scene Text Detector) deep learning model for detecting text in natural scene images. It includes pre-trained models achieving 80.83 F1 on ICDAR 2015, a fast C++ Locality-Aware NMS, and training scripts.

Use cases

• detect text regions in scene images
• run OCR text detection on photos
• train a scene text detector on ICDAR data
• find text bounding boxes in natural images
• build a text detection demo server
• extract text locations from images before OCR

When to choose

• you need fast scene text detection with TensorFlow
• you want a pre-trained EAST model with ICDAR 2015-level accuracy
• you need GPU-accelerated text region detection at ~16 fps on 720p images

When to avoid

• you need the QUAD geometry output (only RBOX is implemented)
• you want a maintained project with recent updates
• you need PyTorch or modern TensorFlow 2.x support
• you need full OCR (recognition) rather than just detection

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision ocr image-processing computer-vision machine-learning deep-learning image-processing python cpp cross-platform tensorflow scene-text-detection east text-detector resnet icdar2015 nms gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="argman/EAST")