Brain-inspired learning to Deeper Inductive Reasoning for Video Captioning (https://doi.org/10.1007/s13042-023-01876-9)
This repo is accepted by "International Journal of Machine Learning and Cybernetics" (IF=5.6 2023.06)
Some files you may need(ckpt,json,datasets...): https://pan.baidu.com/s/1ZraMbY8eEFKNHBgkz_RiUA(rxlq)
The "caption_eval" is also recommended to download in any other repo, cuz it noticed me some files is larger than 50MB and I am not sure it is uploaded successfully.
The "visible.py" can visualize some video samples, so you can run first.
Please notice that we choose the videos in "MSR-VTT". So you should download the "./dataset/MSR-VTT/*", "dict2.pt" and "warm_up.pt".
Just run the visible.py and see what happened !
The code is some kind of mass cuz I tried many times in this experiment.
You can train your own model in MSVD and MSR-VTT, and we do not offer any pretrained models becuase of the mass version control...
If you find our code or our paper useful for your research, please cite our work:
Coming soon.
If you have any question or comment, please contact 1962410334@hhu.edu.cn.
It also should be noticed that I will studied in ZJU soon. So you can contact my personal email 975136225@qq.com.