标签:

• 图像转文本
• 图像描述生成 许可证: Apache-2.0 示例:
• 图片: https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg 示例标题: 热带草原
• 图片: https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg 示例标题: 足球比赛
• 图片: https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg 示例标题: 机场

nlpconnect/vit-gpt2-image-captioning

这是由@ydshieh在flax上训练的图片描述生成模型，此版本为该模型的PyTorch实现。

使用Transformers的图解图像描述生成

• 详细说明见: https://ankur3107.github.io/blogs/the-illustrated-image-captioning-using-transformers/

运行示例代码

from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
import torch
from PIL import Image

model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
feature_extractor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning")

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

max_length = 16
num_beams = 4
gen_kwargs = {"max_length": max_length, "num_beams": num_beams}

def predict_step(image_paths):
    images = []
    for image_path in image_paths:
        i_image = Image.open(image_path)
        if i_image.mode != "RGB":
            i_image = i_image.convert(mode="RGB")
        images.append(i_image)

    pixel_values = feature_extractor(images=images, return_tensors="pt").pixel_values
    pixel_values = pixel_values.to(device)

    output_ids = model.generate(pixel_values, **gen_kwargs)

    preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
    preds = [pred.strip() for pred in preds]
    return preds

predict_step(['doctor.e16ba4e4.jpg']) # 输出: ['一位女士躺在病床上，旁边站着另一位女士']

使用Transformers流水线的示例代码

from transformers import pipeline

image_to_text = pipeline("image-to-text", model="nlpconnect/vit-gpt2-image-captioning")

image_to_text("https://ankur3107.github.io/assets/images/image-captioning-example.png")

# 输出: [{'generated_text': '足球比赛中一名球员跃起接球'}]

获取帮助

• 联系地址: https://huggingface.co/ankur310794
• 推特: https://twitter.com/ankur310794
• GitHub: http://github.com/ankur3107
• 领英: https://www.linkedin.com/in/ankur310794