Videomaev2 Large

🚀 VideoMAE-v2（大型模型，在UnlabeledHybrid-1M上预训练）

VideoMAE-v2 大型模型在 UnlabeldHybrid-1M 数据集上以自监督的方式进行了 800 个 epoch 的预训练。该模型由 Wang 等人在论文 [CVPR23]VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking 中提出，并首次在 GitHub 上发布。它可用于视频特征提取等任务。

🚀 快速开始

你可以使用该原始模型进行视频特征提取。

💻 使用示例

基础用法

from transformers import VideoMAEImageProcessor, AutoModel, AutoConfig
import numpy as np
import torch


config = AutoConfig.from_pretrained("OpenGVLab/VideoMAEv2-Large", trust_remote_code=True)
processor = VideoMAEImageProcessor.from_pretrained("OpenGVLab/VideoMAEv2-Large")
model = AutoModel.from_pretrained('OpenGVLab/VideoMAEv2-Large', config=config, trust_remote_code=True)


video = list(np.random.rand(16, 3, 224, 224))

# B, T, C, H, W -> B, C, T, H, W
inputs = processor(video, return_tensors="pt")
inputs['pixel_values'] = inputs['pixel_values'].permute(0, 2, 1, 3, 4)

with torch.no_grad():
  outputs = model(**inputs)

📚 详细文档

预期用途和限制

你可以使用该原始模型进行视频特征提取。

BibTeX 引用和引用信息

@InProceedings{wang2023videomaev2,
    author    = {Wang, Limin and Huang, Bingkun and Zhao, Zhiyu and Tong, Zhan and He, Yinan and Wang, Yi and Wang, Yali and Qiao, Yu},
    title     = {VideoMAE V2: Scaling Video Masked Autoencoders With Dual Masking},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2023},
    pages     = {14549-14560}
}

@misc{videomaev2,
      title={VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking},
      author={Limin Wang and Bingkun Huang and Zhiyu Zhao and Zhan Tong and Yinan He and Yi Wang and Yali Wang and Yu Qiao},
      year={2023},
      eprint={2303.16727},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

📄 许可证

本项目采用 CC BY-NC 4.0 许可证。