https://github.com/qiantianwen/NuScenes-QA

[AAAI 2024] NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario.
https://github.com/qiantianwen/NuScenes-QA

autonomous-driving vision-language visual-question-answering

Last synced: about 1 month ago
JSON representation

[AAAI 2024] NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario.

• Host: GitHub
• URL: https://github.com/qiantianwen/NuScenes-QA
• Owner: qiantianwen
• License: mit
• Created: 2023-05-24T06:28:17.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-12-10T14:03:35.000Z (7 months ago)
• Last Synced: 2024-02-10T16:32:34.380Z (5 months ago)
• Topics: autonomous-driving, vision-language, visual-question-answering
• Homepage:
• Size: 1.55 MB
• Stars: 101
• Watchers: 14
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• PJLab-ADG-awesome-knowledge-driven-AD - NuScenes-QA
• awesome-knowledge-driven-AD - NuScenes-QA
• Awesome-LLM4AD - NuScenes-QA

README

        
# [AAAI 2024] NuScenes-QA

Official repository for the AAAI 2024 paper **[NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario](https://arxiv.org/pdf/2305.14836.pdf)**.

![DataConstruction](docs/data_construction.png)

## :fire: News

- `2023.12.09`  Our paper is accepted by AAAI 2024! 

- `2023.09.04`  Our NuScenes-QA dataset v1.0 released.

## :hourglass_flowing_sand: To Do

- [x] Release question & anwswer data

- [ ] Release visual feature

- [ ] Release training and testing code

## :running: Getting Started

### Data Preparation

We have released our question-answer annotations, please download it from [HERE](https://drive.google.com/drive/folders/1jIkICT23wZWZYPrWCa0x-ubjpClSzOuU?usp=sharing).

For the visual data, you can download the origin nuScenes dataset from [HERE](https://www.nuscenes.org/download), and prepare the data refer to this [LINK](https://mmdetection3d.readthedocs.io/en/v0.16.0/datasets/nuscenes_det.html). As an alternative, you can also download our provided object-level features extracted using pre-trained detection models from [HERE]() (to be released soon).

### Training & Testing

Todo.

## :star: Others

If you have any questions about the dataset and its generation or the object-level feature extraction, feel free to cantact me with `[email protected]`.

## :book: Citation

If you find our paper and project useful, please consider citing:

```bibtex

@article{qian2023nuscenes,

  title={NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario},

  author={Qian, Tianwen and Chen, Jingjing and Zhuo, Linhai and Jiao, Yang and Jiang, Yu-Gang},

  journal={arXiv preprint arXiv:2305.14836},

  year={2023}

}

```