https://github.com/notai-tech/keras-craft

Extremely easy to use Text Detection module with CRAFT pre-trained model.
https://github.com/notai-tech/keras-craft

craft cv east keras scene-text-detection scene-text-recognition tensorflow text-detection text-recognition

Last synced: 3 days ago
JSON representation

Extremely easy to use Text Detection module with CRAFT pre-trained model.

• Host: GitHub
• URL: https://github.com/notai-tech/keras-craft
• Owner: notAI-tech
• License: mit
• Created: 2020-01-23T10:53:44.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2022-11-21T21:40:15.000Z (almost 2 years ago)
• Last Synced: 2023-03-10T10:44:07.938Z (over 1 year ago)
• Topics: craft, cv, east, keras, scene-text-detection, scene-text-recognition, tensorflow, text-detection, text-recognition
• Language: Python
• Homepage:
• Size: 24.4 KB
• Stars: 6
• Watchers: 2
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# keras-craft

Extremely easy to use Text Detection module with CRAFT pre-trained model.

keras-craft aims to be production ready and supports features like batch inference (auto batching for images of different size) and tensorflow serving.

# Installation

```pip install git+https://github.com/notAI-tech/keras-craft``` (the entire library)

# Usage (craft_client)

```bash

docker run -p 8500:8500 bedapudi6788/keras-craft:generic-english

```

```python

import craft_client

image_paths = [image_1, image_2, ..]

all_boxes = craft_client.detect(image_paths)

# Visualization

for image_path, boxes in zip(image_paths):

  image_with_boxes_path = craft_client.draw_boxes_on_image(image_path, boxes)

  print(image_with_boxes_path)

```

# Usage (keras_craft)

```python

import keras_craft

detector = keras_craft.Detector()

image_paths = [image_1, image_2, ..]

all_boxes = detector.detect(image_paths)

# Visualization

for image_path, boxes in zip(image_paths):

  image_with_boxes_path = keras_craft.draw_boxes_on_image(image_path, boxes)

  print(image_with_boxes_path)

```

# Example image_with_boxes

![](https://i.imgur.com/EtGvyCz.png)

# To Do:

1. Train different models for different use-cases. (various languages ..)

2. Experiment with smaller model(s)

**Credit for the core keras model, generic-english checkpoint .. goes to [Fausto Morales](https://github.com/faustomorales/keras-ocr) and Clova.ai**