init

app.py

@@ -0,0 +1,45 @@
+# import os
+# os.system('pip install git+https://github.com/huggingface/transformers.git --upgrade')
+
+import gradio as gr
+from transformers import TrOCRProcessor, VisionEncoderDecoderModel
+import requests
+from PIL import Image
+
+processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-printed")
+model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-printed")
+model.config.eos_token_id = 2
+
+# load image examples
+urls = ['https://layoutlm.blob.core.windows.net/trocr/dataset/SROIE2019Task2Crop/train/X00016469612_1.jpg', 'https://layoutlm.blob.core.windows.net/trocr/dataset/SROIE2019Task2Crop/train/X51005255805_7.jpg', 'https://layoutlm.blob.core.windows.net/trocr/dataset/SROIE2019Task2Crop/train/X51005745214_6.jpg']
+for idx, url in enumerate(urls):
+  image = Image.open(requests.get(url, stream=True).raw)
+  image.save(f"image_{idx}.png")
+
+def process_image(image):
+    # prepare image
+    pixel_values = processor(image, return_tensors="pt").pixel_values
+
+    # generate (no beam search)
+    generated_ids = model.generate(pixel_values)
+
+    # decode
+    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
+
+    return generated_text
+
+title = "Interactive demo: TrOCR"
+description = "Demo for Microsoft's TrOCR, an encoder-decoder model consisting of an image Transformer encoder and a text Transformer decoder for state-of-the-art optical character recognition (OCR) on single-text line images. This particular model is fine-tuned on SROIE Task 2, a dataset of annotated printed images. To use it, simply upload a (single-text line) image or use one of the example images below and click 'submit'. Results will show up in a few seconds."
+article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2109.10282'>TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models</a> | <a href='https://github.com/microsoft/unilm/tree/master/trocr'>Github Repo</a></p>"
+examples =[["image_0.png"], ["image_1.png"], ["image_2.png"]]
+
+#css = """.output_image, .input_image {height: 600px !important}"""
+
+iface = gr.Interface(fn=process_image, 
+                     inputs=gr.inputs.Image(type="pil"), 
+                     outputs=gr.outputs.Textbox(),
+                     title=title,
+                     description=description,
+                     article=article,
+                     examples=examples)
+iface.launch(debug=True)

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+torch
+Pillow
+transformers