pipeline_tag: 图像转文本 tags:

• 图像描述生成 languages:
• 英文 license: bsd-3-clause

import requests
from PIL import Image
from transformers import BlipProcessor, BlipForConditionalGeneration
processor = BlipProcessor.from_pretrained("Revrse/icon-captioning-model")
model = BlipForConditionalGeneration.from_pretrained("Revrse/icon-captioning-model")
img_url = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/demo.jpg' 
raw_image = Image.open(requests.get(img_url, stream=True).raw).convert('RGB')
# 条件式图像描述生成
text = "一张摄影作品，内容是"
inputs = processor(raw_image, text, return_tensors="pt")
out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))
# 无条件式图像描述生成
inputs = processor(raw_image, return_tensors="pt")
out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))