BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation

在 COCO 数据集上进行预训练的图像描述模型卡 - 基础架构（具有 ViT 大型骨干网络）。


从BLIP官方仓库中提取数据

摘要

作者在论文的摘要中写道：

视觉-语言预训练（VLP）提高了许多视觉-语言任务的性能上取得了进展。然而，大多数现有的预训练模型仅在理解任务或生成任务中表现出色。此外，通过使用从网络收集的包含噪声的图像-文本对来扩展数据集，在很大程度上实现了性能的提高，这是一个次优的监督来源。在本文中，我们提出了 BLIP，一种新的 VLP 框架，它可以灵活地应用于视觉-语言理解和生成任务。BLIP 通过自展标注（bootstrapping the captions），可以有效地利用带有噪声的 web 数据，其中标注器（captioner）生成标注，过滤器（filter）去除有噪声的标注。该研究在视觉 - 语言任务上取得了 SOTA 性能，例如在图像 - 文本检索任务上， recall@1 提高 2.7%；在图像标注任务上，CIDEr 提高 2.8%、VQA 提高 +1.6%。当将 BLIP 以零样本的方式直接迁移到视频 - 语言任务时，BLIP 也表现出很强的泛化能力。代码、模型和数据集已发布。

用途

您可以使用此模型进行有条件和无条件的图像描述生成。

使用 Pytorch 模型

在 CPU 上运行模型

import requests
from PIL import Image
from transformers import BlipProcessor, BlipForConditionalGeneration

processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")

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')

# conditional image captioning
text = "a photography of"
inputs = processor(raw_image, text, return_tensors="pt")

out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))

# unconditional image captioning
inputs = processor(raw_image, return_tensors="pt")

out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))

在 GPU 上运行模型

全精度

import requests
from PIL import Image
from transformers import BlipProcessor, BlipForConditionalGeneration

processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large").to("cuda")

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')

# conditional image captioning
text = "a photography of"
inputs = processor(raw_image, text, return_tensors="pt").to("cuda")

out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))

# unconditional image captioning
inputs = processor(raw_image, return_tensors="pt").to("cuda")

out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))

半精度 (`float16`)

import torch
import requests
from PIL import Image
from transformers import BlipProcessor, BlipForConditionalGeneration

processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large", torch_dtype=torch.float16).to("cuda")

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')

# conditional image captioning
text = "a photography of"
inputs = processor(raw_image, text, return_tensors="pt").to("cuda", torch.float16)

out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))
# >>> a photography of a woman and her dog

# unconditional image captioning
inputs = processor(raw_image, return_tensors="pt").to("cuda", torch.float16)

out = model.generate(**inputs)
print(processor.decode(out[0], skip_special_tokens=True))
>>> a woman sitting on the beach with her dog

BibTex条目和引用信息

@misc{https://doi.org/10.48550/arxiv.2201.12086,
  doi = {10.48550/ARXIV.2201.12086},
  
  url = {https://arxiv.org/abs/2201.12086},
  
  author = {Li, Junnan and Li, Dongxu and Xiong, Caiming and Hoi, Steven},
  
  keywords = {Computer Vision and Pattern Recognition (cs.CV), FOS: Computer and information sciences, FOS: Computer and information sciences},
  
  title = {BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation},
  
  publisher = {arXiv},
  
  year = {2022},
  
  copyright = {Creative Commons Attribution 4.0 International}
}

15 KiB Raw Blame History Unescape Escape