Spaces:
Running
Running
File size: 8,802 Bytes
5bedb5a |
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 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
# coding: utf-8
# Copyright (C) 2023, [Breezedeus](https://github.com/breezedeus).
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
# Ref: https://huggingface.co/spaces/hysts/Manga-OCR/blob/main/app.py
import os
import json
import functools
import random
import string
import time
import yaml
import gradio as gr
import numpy as np
# from cnstd.utils import pil_to_numpy, imsave
from pix2text import Pix2Text
from pix2text.utils import set_logger, merge_line_texts
logger = set_logger()
LANGUAGES = yaml.safe_load(open('languages.yaml', 'r', encoding='utf-8'))['languages']
def get_p2t_model(lan_list: list):
p2t = Pix2Text(languages=lan_list)
return p2t
def latex_render(latex_str):
return f"$$\n{latex_str}\n$$"
# return latex_str
def recognize(lang_list, rec_type, resized_shape, image_file):
lang_list = [LANGUAGES[l] for l in lang_list]
p2t = get_p2t_model(lang_list)
if rec_type == 'Formula & Text':
suffix = list(string.ascii_letters)
random.shuffle(suffix)
suffix = ''.join(suffix[:6])
out_det_fp = f'out-det-{time.time()}-{suffix}.jpg'
outs = p2t(
image_file, resized_shape=resized_shape, save_analysis_res=out_det_fp
)
# To get just the text contents, use:
only_text = merge_line_texts(outs, auto_line_break=True)
# return only_text, latex_render(only_text)
return only_text, out_det_fp
elif rec_type == 'Only Formula':
only_text = p2t.recognize_formula(image_file)
return latex_render(only_text), None
elif rec_type == 'Only Text':
only_text = p2t.recognize_text(image_file)
return only_text, None
def main():
langs = list(LANGUAGES.keys())
langs.sort(key=lambda x: x.lower())
title = 'Demo'
# example_func = functools.partial(
# recognize,
# new_size=768,
# box_score_thresh=0.3,
# min_box_size=10,
# )
# examples = [
# [
# 'ch_PP-OCRv3_det::onnx',
# True,
# 'number-densenet_lite_136-fc',
# False,
# 'docs/examples/card1-s.jpg',
# ],
# [
# 'ch_PP-OCRv3_det::onnx',
# True,
# 'number-densenet_lite_136-fc',
# False,
# 'docs/examples/card2-s.jpg',
# ],
# [
# 'ch_PP-OCRv3_det::onnx',
# True,
# 'number-densenet_lite_136-fc',
# False,
# 'docs/examples/cy1-s.jpg',
# ],
# [
# 'ch_PP-OCRv3_det::onnx',
# False,
# 'densenet_lite_136-gru',
# False,
# 'docs/examples/huochepiao.jpeg',
# ],
# [
# 'ch_PP-OCRv3_det::onnx',
# False,
# 'densenet_lite_136-gru',
# False,
# 'docs/examples/1_res.jpg',
# ],
# [
# 'db_shufflenet_v2::pytorch',
# False,
# 'en_number_mobile_v2.0',
# False,
# 'docs/examples/en_book1.jpeg',
# ],
# [
# 'db_shufflenet_v2::pytorch',
# False,
# 'densenet_lite_136-gru',
# True,
# 'docs/examples/beauty0.jpg',
# ],
# ]
table_desc = """
<div align="center">
<img src="https://www.notion.so/image/https%3A%2F%2Fprod-files-secure.s3.us-west-2.amazonaws.com%2F9341931a-53f0-48e1-b026-0f1ad17b457c%2Fc41e0b1d-4869-4e39-93db-631569e6a38d%2FUntitled.png?table=block&id=3d0819ca-2e1a-46a7-b6f3-b4cf89cd045c" width="120px"/>
[](https://visitorbadge.io/status?path=https%3A%2F%2Fhuggingface.co%2Fspaces%2Fbreezedeus%2FCnOCR-Demo)
[](https://discord.gg/H9FmDSMA)
| | |
| ------------------------------- | --------------------------------------- |
| 🏄 **Free Web Service** | [p2t.breezedeus.com](https://p2t.breezedeus.com) |
| 📀 **Code** | [Github](https://github.com/breezedeus/pix2text) |
| 💬 **Discord** | [P2T @ Discord](https://discord.gg/H9FmDSMA) |
| 👨🏻💻 **Author** | [Breezedeus](https://www.breezedeus.com) |
If useful, please help to **star 🌟 [Pix2Text](https://github.com/breezedeus/pix2text)** 🙏
</div>
"""
with gr.Blocks() as demo:
gr.HTML(
f'<h1 style="text-align: center; margin-bottom: 1rem;"><a href="https://github.com/breezedeus/pix2text" target="_blank">Pix2Text V1.0</a> {title}</h1>'
)
with gr.Row(equal_height=False):
with gr.Column(min_width=200, variant='panel', scale=3):
gr.Markdown('### Settings')
lang_list = gr.Dropdown(
label='Text Languages',
choices=langs,
value=['English', 'Chinese Simplified'],
multiselect=True,
info='Which languages to be recognized as Texts.',
)
rec_type = gr.Radio(
choices=['Formula & Text', 'Only Formula', 'Only Text'],
label='Image Type',
value='Formula & Text',
info='Which type of image to be recognized.',
)
resized_shape = gr.Slider(
label='resized_shape',
minimum=512,
maximum=2048,
value=608,
step=32,
)
# with gr.Accordion('Choose Text Languages', open=False):
# lang_list = gr.Checkboxgroup(
# label='Text Languages',
# choices=langs,
# value=['English', 'Chinese Simplified'],
# )
with gr.Column(scale=6, variant='compact'):
gr.Markdown('### Upload Image to be Recognized')
image_file = gr.Image(label='Image', type="pil", image_mode='RGB', show_label=False)
sub_btn = gr.Button("Submit", variant="primary")
with gr.Column(scale=2, variant='compact'):
gr.Markdown(table_desc)
with gr.Row(equal_height=False):
with gr.Column(scale=1, variant='compact'):
gr.Markdown('**Detection Result**')
det_result = gr.Image(
label='Detection Result', scale=1, show_label=False
)
with gr.Column(scale=1, variant='compact'):
gr.Markdown('**Recognition Result**')
rec_result = gr.Textbox(
label=f'Recognition Result',
lines=5,
value='',
scale=1,
show_label=False,
show_copy_button=True,
)
# render_result = gr.Markdown(label=f'After Rendering', value='')
# rec_result.change(latex_render, rec_result, render_result)
sub_btn.click(
recognize,
inputs=[lang_list, rec_type, resized_shape, image_file,],
outputs=[rec_result, det_result],
)
# gr.Examples(
# label='示例',
# examples=examples,
# inputs=[
# det_model_name,
# is_single_line,
# rec_model_name,
# use_angle_clf,
# image_file,
# ],
# outputs=[out_image, naive_warn, out_texts],
# fn=example_func,
# cache_examples=os.getenv('CACHE_EXAMPLES') == '1',
# )
demo.queue(max_size=10)
demo.launch()
if __name__ == '__main__':
main()
|