Skip to content

ZJCV/Non-local

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

语言: 🇨🇳 🇺🇸

«Non-local»复现了论文Non-local Neural Networks提出的视频分类模型

内容列表

背景

受到传统图像处理算法non-local means影响,Non-local Neural Networks提出了注意力模块Non-local Block

安装

通过requirements.txt安装运行所需依赖

$ pip install -r requirements.txt

处理数据时需要额外安装denseflow,可以在innerlee/setup中找到安装脚本

使用

首先设置GPU和当前位置

$ export CUDA_VISIBLE_DEVICES=1
$ export PYTHONPATH=.
  • 训练
# 训练UCF101
# 单GPU
$ python tools/train.py --config_file=configs/i3d-3x3-nl_r3d50_ucf101_rgb_224x32_dense.yaml
# 多GPU
$ python tools/train.py \
--config_file=configs/i3d-3x3-nl_r3d50_ucf101_rgb_224x32_dense.yaml \
--eval_step=1000 \
--save_step=1000 \
-g=<N>
  • 测试
$ python tools/test.py <config_file> <pth_file>
$ python tools/test.py configs/i3d-3x3-nl_r3d50_ucf101_rgb_224x32_dense.yaml outputs/i3d-3x3-nl_r3d50_ucf101_rgb_224x32_dense.pth

主要维护人员

  • zhujian - Initial work - zjykzj

致谢

@misc{fan2020pyslowfast,
  author =       {Haoqi Fan and Yanghao Li and Bo Xiong and Wan-Yen Lo and
                  Christoph Feichtenhofer},
  title =        {PySlowFast},
  howpublished = {\url{https://github.com/facebookresearch/slowfast}},
  year =         {2020}
}


@misc{wang2018nonlocal,
      title={Non-local Neural Networks}, 
      author={Xiaolong Wang and Ross Girshick and Abhinav Gupta and Kaiming He},
      year={2018},
      eprint={1711.07971},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

参与贡献方式

欢迎任何人的参与!打开issue或提交合并请求。

注意:

许可证

Apache License 2.0 © 2020 zjykzj