https://github.com/sutdcv/SUTD-TrafficQA

[CVPR2021] SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning over Traffic Events
https://github.com/sutdcv/SUTD-TrafficQA

annotations cvpr cvpr2021 dataset multimodal multimodal-deep-learning paper traffic-events video-qa video-reasoning vqa vqa-dataset

Last synced: 16 days ago
JSON representation

[CVPR2021] SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning over Traffic Events

• Host: GitHub
• URL: https://github.com/sutdcv/SUTD-TrafficQA
• Owner: sutdcv
• Created: 2021-03-27T11:20:40.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2024-08-19T04:17:43.000Z (3 months ago)
• Last Synced: 2024-08-19T05:29:40.897Z (3 months ago)
• Topics: annotations, cvpr, cvpr2021, dataset, multimodal, multimodal-deep-learning, paper, traffic-events, video-qa, video-reasoning, vqa, vqa-dataset
• Language: JavaScript
• Homepage: https://sutdcv.github.io/SUTD-TrafficQA/
• Size: 6 MB
• Stars: 49
• Watchers: 4
• Forks: 2
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • Citation: CITATION

Awesome Lists containing this project

• Awesome-Multimodal-LLM-Autonomous-Driving - SUTD-TrafficQA 2021 - Collected | QA | QA | 10k frames 62k text strings | **:heavy_check_mark:** | (Awesome Papers / Datasets)
• Awesome-TimeSeries-SpatioTemporal-LM-LLM - \[link\

README

        
# [SUTD-TrafficQA](https://sutdcv.github.io/SUTD-TrafficQA/)

A challenging **Video Question Answering (VQA)** Benchmark based on **real-world** traffic scenes.

**Updates:**

-   `Jul 2021` The dataset is publicly released. You may [request download](https://sutdcv.github.io/SUTD-TrafficQA/#/download) now.

-   `Jun 2021` The [dataset usage details](https://sutdcv.github.io/SUTD-TrafficQA/#/explore) are available now.

-   `May 2021` The dataset [homepage](https://sutdcv.github.io/SUTD-TrafficQA) is live now.

-   `Feb 2021` ~~The dataset is available upon email request.~~

## Paper

Our paper at **CVPR 2021**, _SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning over Traffic Events_, is available at: [[CVF Open Access]](https://openaccess.thecvf.com/content/CVPR2021/html/Xu_SUTD-TrafficQA_A_Question_Answering_Benchmark_and_an_Efficient_Network_for_CVPR_2021_paper.html), [[arXiv:2103.15538]](https://arxiv.org/abs/2103.15538), and [[ResearchGate]](https://www.researchgate.net/publication/350432154_TrafficQA_A_Question_Answering_Benchmark_and_an_Efficient_Network_for_Video_Reasoning_over_Traffic_Events).

## Dataset

-   Annotation Example [examples/annotation_sample.jsonl](examples/annotation_sample.jsonl)

-   Jsonl Reader Example [examples/jsonl_reader.py](examples/jsonl_reader.py)

-   Appearance Feature Preprocessing [examples/preprocess_video_appearance_example.py](examples/preprocess_video_appearance_example.py)

-   Motion Feature Preprocessing [examples/preprocess_video_motion_example.py](examples/preprocess_video_motion_example.py)

-   Dataloader [examples/dataloader_example.py](examples/dataloader_example.py)

-   [Download Dataset](https://sutdcv.github.io/SUTD-TrafficQA)

## Citation

```bibtex

@InProceedings{Xu_2021_CVPR,

    author    = {Xu, Li and Huang, He and Liu, Jun},

    title     = {{SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning Over Traffic Events}},

    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},

    month     = {June},

    year      = {2021},

    pages     = {9878-9888}

}

```

## Acknowledgment

Contributors: [**Lin Yutian**](https://github.com/Lynn-020809), [**Tran Nguyen Bao Long**](https://github.com/TNBL265), [**Liu Renhang**](https://github.com/Samillynn), [**Qiao Yingjie**](https://github.com/YingjieQiao), **Xun Long Ng**, **Koh Kai Ting**, **Christabel Dorothy**

Code Reference: [thaolmk54 / hcrn-videoqa](https://github.com/thaolmk54/hcrn-videoqa)

## Contact

-   `li_xu [AT] mymail.sutd.edu.sg`

-   `he_huang [AT] mymail.sutd.edu.sg`