Spaces:
Runtime error
Runtime error
File size: 2,444 Bytes
2773523 3a227f4 2773523 075ee8a 2773523 a808d84 2773523 38ab1e3 2773523 a808d84 2773523 3a227f4 2773523 329d18e 2773523 329d18e 2773523 3a227f4 2773523 |
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 |
import gradio as gr
from transformers import AutoProcessor, AutoModelForCausalLM, BlipForConditionalGeneration
import torch
torch.hub.download_url_to_file('http://images.cocodataset.org/val2017/000000039769.jpg', 'cats.jpg')
torch.hub.download_url_to_file('https://huggingface.co/datasets/nielsr/textcaps-sample/resolve/main/stop_sign.png', 'stop_sign.png')
git_processor = AutoProcessor.from_pretrained("microsoft/git-base-coco")
git_model = AutoModelForCausalLM.from_pretrained("microsoft/git-base-coco")
blip_processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
blip_model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
device = "cuda" if torch.cuda.is_available() else "cpu"
git_model.to(device)
blip_model.to(device)
def generate_caption(processor, model, image):
inputs = processor(images=image, return_tensors="pt").to(device)
generated_ids = model.generate(pixel_values=inputs.pixel_values, max_length=50)
generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
return generated_caption
def generate_captions(image):
caption_git = generate_caption(git_processor, git_model, image)
caption_blip = generate_caption(blip_processor, blip_model, image)
return caption_git, caption_blip
examples = [["cats.jpg", "stop_sign.png"]]
title = "Interactive demo: comparing image captioning models"
description = "Gradio Demo to compare GIT and BLIP, 2 state-of-the-art captioning models. To use it, simply upload your image and click 'submit', or click one of the examples to load them. Read more at the links below."
article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2102.03334' target='_blank'>ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision</a> | <a href='https://github.com/dandelin/ViLT' target='_blank'>Github Repo</a></p>"
interface = gr.Interface(fn=generate_captions,
inputs=gr.inputs.Image(type="pil"),
outputs=[gr.outputs.Textbox(label="Caption generated by GIT"), gr.outputs.Textbox(label="Caption generated by BLIP")],
examples=examples,
title=title,
description=description,
article=article,
enable_queue=True)
interface.launch(debug=True) |