Spaces:
Runtime error
Runtime error
File size: 3,913 Bytes
e47538b cf6b359 e47538b cf6b359 e47538b 75f9e87 e47538b cf6b359 266e7e3 2adbf8c e47538b 99a4b22 56cdfa2 e47538b 80f1479 e47538b 99a4b22 e47538b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
import pandas as pd
import PIL
from PIL import Image
from PIL import ImageDraw
import gradio as gr
import torch
import easyocr
import omegaconf
from vietocr.model.transformerocr import VietOCR
from vietocr.model.vocab import Vocab
from vietocr.translate import translate, process_input
config = omegaconf.OmegaConf.load("vgg-seq2seq.yaml")
config = omegaconf.OmegaConf.to_container(config, resolve=True)
vocab = Vocab(config['vocab'])
model = VietOCR(len(vocab),
config['backbone'],
config['cnn'],
config['transformer'],
config['seq_modeling'])
model.load_state_dict(torch.load('train_old.pth', map_location=torch.device('cpu')))
torch.hub.download_url_to_file('https://github.com/JaidedAI/EasyOCR/raw/master/examples/english.png', 'english.png')
torch.hub.download_url_to_file('https://github.com/JaidedAI/EasyOCR/raw/master/examples/thai.jpg', 'thai.jpg')
torch.hub.download_url_to_file('https://github.com/JaidedAI/EasyOCR/raw/master/examples/french.jpg', 'french.jpg')
torch.hub.download_url_to_file('https://github.com/JaidedAI/EasyOCR/raw/master/examples/chinese.jpg', 'chinese.jpg')
torch.hub.download_url_to_file('https://github.com/JaidedAI/EasyOCR/raw/master/examples/japanese.jpg', 'japanese.jpg')
torch.hub.download_url_to_file('https://github.com/JaidedAI/EasyOCR/raw/master/examples/korean.png', 'korean.png')
torch.hub.download_url_to_file('https://i.imgur.com/mwQFd7G.jpeg', 'Hindi.jpeg')
def draw_boxes(image, bounds, color='yellow', width=2):
draw = ImageDraw.Draw(image)
for bound in bounds:
p0, p1, p2, p3 = bound[0]
draw.line([*p0, *p1, *p2, *p3, *p0], fill=color, width=width)
return image
def inference(filepath, lang):
reader = easyocr.Reader(lang)
bounds = reader.readtext(filepath)
new_bounds=[]
for (bbox, text, prob) in bounds:
x1,y1 = bbox[0]
x2,y2 = bbox[1]
x3,y3 = bbox[2]
x4,y4 = bbox[3]
# crop the region of interest (ROI)
img = Image.open(filepath)
#img = img[y0:y1, x0:x1]
img = img.crop((max(0,x1-5), max(0,y1-5), min(x3+5,width), min(y3+5, height))) # crop the image
img = process_input(img, config['dataset']['image_height'],
config['dataset']['image_min_width'], config['dataset']['image_max_width'])
out = translate(img, model)[0].tolist()
out = vocab.decode(out)
new_bounds.append((bbox,text, out, prob))
im = PIL.Image.open(filepath)
draw_boxes(im, bounds)
im.save('result.jpg')
return ['result.jpg', pd.DataFrame(bounds).iloc[: , 1:]]
title = 'EasyOCR'
description = 'Gradio demo for EasyOCR. EasyOCR demo supports 80+ languages.To use it, simply upload your image and choose a language from the dropdown menu, or click one of the examples to load them. Read more at the links below.'
article = "<p style='text-align: center'><a href='https://www.jaided.ai/easyocr/'>Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin, Chinese, Arabic, Devanagari, Cyrillic and etc.</a> | <a href='https://github.com/JaidedAI/EasyOCR'>Github Repo</a></p>"
examples = [['english.png',['en']],['thai.jpg',['th']],['french.jpg',['fr', 'en']],['chinese.jpg',['ch_sim', 'en']],['japanese.jpg',['ja', 'en']],['korean.png',['ko', 'en']],['Hindi.jpeg',['hi', 'en']]]
css = ".output_image, .input_image {height: 40rem !important; width: 100% !important;}"
choices = [
"vi"
]
gr.Interface(
inference,
[gr.inputs.Image(type='filepath', label='Input'),gr.inputs.CheckboxGroup(choices, type="value", default=['vi'], label='language')],
[gr.outputs.Image(type='pil', label='Output'), gr.outputs.Dataframe(type='pandas', headers=['easyOCR','vietOCR', 'confidence'])],
title=title,
description=description,
article=article,
examples=examples,
css=css,
enable_queue=True
).launch(debug=True) |