Skip to content
The project page of paper: Learning Neural Bag-of-Matrix-Summarization with Riemannian Network [AAAI 2019]
Python
Branch: master
Clone or download

Latest commit

Fetching latest commit…
Cannot retrieve the latest commit at this time.

Files

Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
BiMap.py
DisNet.py
LogEig.py
README.md
ReEig.py
RieDictionaryLayer.py
main.py
mat.py
model.py
myutil.py
spdnet.py

README.md

This is the project page of our paper:

Learning Neural Bag-of-Matrix-Summarization with Riemannian Network,
Liu, H., Li, J., Wu, Y., & Ji, R. AAAI 2019. [PDF]

Code

Check main.py for details.

Our codes are based on zzhiwu/SPDNet, and translated into PyTorch.

Different from the original paper, we find that replacing the classifier with the softmax on the distances between the dictionary is more stable.

Note that best of results on AFEW is fine-tuned from the SPDNet.

Feel free to contact to the authors (lijie.32@outlook.com) if you have any problems.

Citation

If our paper helps your research, please cite it in your publications:

@InProceedings{liu2019learning,
  title={Learning Neural Bag-of-Matrix-Summarization with Riemannian Network},
  author={Liu, Hong and Li, Jie and Wu, Yongjian and Ji, Rongrong},
  booktitle={Association for the Advancement of Artificial Intelligence (AAAI)},
  year={2019}
}

@InProceedings{huang2017riemannian,
  title={A riemannian network for spd matrix learning},
  author={Huang, Zhiwu and Van Gool, Luc},
  booktitle={Thirty-First AAAI Conference on Artificial Intelligence},
  year={2017}
}
You can’t perform that action at this time.