| Name | UID |
|---|---|
| Rakesh Bal | 605526216 |
| Sri Keerthi Bolli | 505525868 |
This repository is a PyTorch and Python implementation of the course project for UCLA CS 269 Sem 5 (Human Centred Artificial Intelligence) S22 Course.
We use different exisitng works for our project and we recommend installing them from their respective repositories.
The scripts are provided in this repository for running our experiments are changes we made to various files in the existing projects mentioned above. For example for MCAN-VQA, we recommend cloining the original repo git clone https://github.com/MILVLG/mcan-vqa.git and then replacing the original files with the files provided in the repository under mcan-vqa folder. Similarly, LSeg we provide the changes we made to the lseg_demo.ipynb in the original repository. For multilingual vqa, the scipt provided converts the multilingual files provided in the dataset (here) to the corresponding jsons to be used by MCAN model.
@inProceedings{yu2019mcan,
author = {Yu, Zhou and Yu, Jun and Cui, Yuhao and Tao, Dacheng and Tian, Qi},
title = {Deep Modular Co-Attention Networks for Visual Question Answering},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {6281--6290},
year = {2019}
}
@inproceedings{
li2022languagedriven,
title={Language-driven Semantic Segmentation},
author={Boyi Li and Kilian Q Weinberger and Serge Belongie and Vladlen Koltun and Rene Ranftl},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=RriDjddCLN}
}
@inProceedings{yu2019mcan,
author = {Yu, Zhou and Yu, Jun and Cui, Yuhao and Tao, Dacheng and Tian, Qi},
title = {Deep Modular Co-Attention Networks for Visual Question Answering},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {6281--6290},
year = {2019}
}