# Residual Attention-based LSTM for Video Caption
An implementation for paper "Residual Attention-based LSTM for Video Caption": https://link.springer.com/article/10.1007%2Fs11280-018-0531-z
## Requirements
Python 2.7.6
Theano 0.8.2
## processed data
You need to download pretrained resnet model for extracting features.
We provide our extracted ResNet video feature and processed caption in:https://drive.google.com/open?id=1HymvVvAEygM6UJm41dQkQ4IbTWcHT0iQ. Download this dataset and replace RAB_FEATURE_BASE_PATH in config.py with your feature path and replace RAB_DATASET_BASE_PATH in config.py with your processed data path. Besides, you should assign where to store your result in config.py.
## Evaluation
If you'd like to evaluate BLEU/METEOR/CIDER scores during training. Don't forget
to download coco-caption:https://github.com/tylin/coco-caption and Jobman:http://deeplearning.net/software/jobman/install.html.
Also you should add coco-caption path to $PYTHONPATH and add jobman path to $PYTHONPATH as well.
## Others
If you have any questions, drop us email at:xiangpengli.cs@gmail.com
```
@article{li2019residual,
title={Residual attention-based LSTM for video captioning},
author={Li, Xiangpeng and Zhou, Zhilong and Chen, Lijiang and Gao, Lianli},
journal={World Wide Web},
volume={22},
number={2},
pages={621--636},
year={2019},
publisher={Springer}
}
```
基于残差注意力的 LSTM 视频字幕识别.zip
版权申诉
92 浏览量
2024-03-07
23:02:03
上传
评论
收藏 67KB ZIP 举报
小码蚁.
- 粉丝: 2584
- 资源: 4344
最新资源
- android开发期末大作业基于Androidstudio的医疗系统app源码(高分项目).zip
- assets_e4b6f25e1c38b56b464c56c31c1a361d.mp4
- 基于Android studio设计的图书借阅管理系统APP期末大作业(高分项目).zip
- 安卓期末大作业-音乐播放器App-AndroidStudio开发(高分项目)
- 行人重识别-用于行人重识别的稀疏标签平滑正则化优化-附项目源码+流程教程-优质项目实战.zip
- Python绘图艺术.zip
- 安卓期末大作业(AndroidStudio开发),记事本app源码(高分项目).zip
- 安卓期末大作业Androidstudio-记事本app源码(95分以上).zip
- Java数据结构实现之Queue.zip
- Java数据结构实现之Stack.zip
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈