toloka/WSDMCup2023

---
annotations_creators:
- crowdsourced
language:
- en
language_creators:
- crowdsourced
license:
- cc-by-4.0
multilinguality:
- monolingual
pretty_name: WSDMCup2023
size_categories:
- 10K<n<100K
source_datasets: []
tags:
- toloka
task_categories:
- visual-question-answering
task_ids:
- visual-question-answering
dataset_info:
features:
- name: image
dtype: string
- name: width
dtype: int64
- name: height
dtype: int64
- name: left
dtype: int64
- name: top
dtype: int64
- name: right
dtype: int64
- name: bottom
dtype: int64
- name: question
dtype: string
splits:
- name: train
num_examples: 38990
- name: train_sample
num_examples: 1000
- name: test_public
num_examples: 1705
- name: test_private
num_examples: 4504
config_name: wsdmcup2023
---

# Dataset Card for WSDMCup2023

## Dataset Description

- **Homepage:** [Toloka Visual Question Answering Challenge](https://toloka.ai/challenges/wsdm2023)
- **Repository:** [WSDM Cup 2023 Starter Pack](https://github.com/Toloka/WSDMCup2023)
- **Paper:** <https://arxiv.org/abs/2309.16511>
- **Leaderboard:** [CodaLab Competition Leaderboard](https://codalab.lisn.upsaclay.fr/competitions/7434#results)
- **Point of Contact:** [e-mail]

| Question | Image and Answer |
| --- | --- |
| What do you use to hit the ball? | <img src="https://tlkfrontprod.azureedge.net/portal-production/static/uploaded/images/KUsGAc_eqdMcNxkBXzzl/KUsGAc_eqdMcNxkBXzzl_webp_1280_x2.webp" width="228" alt="What do you use to hit the ball?"> |
| What do people use for cutting? | <img src="https://tlkfrontprod.azureedge.net/portal-production/static/uploaded/images/brXEVYckNLfQKcfNu4DF/brXEVYckNLfQKcfNu4DF_webp_1280_x2.webp" width="228" alt="What do people use for cutting?"> |
| What do we use to support the immune system and get vitamin C? | <img src="https://tlkfrontprod.azureedge.net/portal-production/static/uploaded/images/HQ0A-ZvZCGCmYfTs83K7/HQ0A-ZvZCGCmYfTs83K7_webp_1280_x2.webp" width="228" alt="What do we use to support the immune system and get vitamin C?"> |

### Dataset Summary

The WSDMCup2023 Dataset consists of images associated with textual questions.
One entry (instance) in our dataset is a question-image pair labeled with the ground truth coordinates of a bounding box containing
the visual answer to the given question. The images were obtained from a CC BY-licensed subset of the Microsoft Common Objects in
Context dataset, [MS COCO](https://cocodataset.org/). All data labeling was performed on the [Toloka crowdsourcing platform](https://toloka.ai/).

Our dataset has 45,199 instances split among three subsets: train (38,990 instances), public test (1,705 instances),
and private test (4,504 instances). The entire train dataset was available for everyone since the start of the challenge.
The public test dataset was available since the evaluation phase of the competition but without any ground truth labels.
After the end of the competition, public and private sets were released.

## Dataset Citation

Please cite the challenge results or dataset description as follows.

- Ustalov D., Pavlichenko N., Koshelev S., Likhobaba D., and Smirnova A. [Toloka Visual Question Answering Benchmark](https://arxiv.org/abs/2309.16511). 2023. arXiv: [2309.16511 [cs.CV]](https://arxiv.org/abs/2309.16511).

```bibtex
@inproceedings{TolokaWSDMCup2023,
author = {Ustalov, Dmitry and Pavlichenko, Nikita and Koshelev, Sergey and Likhobaba, Daniil and Smirnova, Alisa},
title = {{Toloka Visual Question Answering Benchmark}},
year = {2023},
eprint = {2309.16511},
eprinttype = {arxiv},
eprintclass = {cs.CV},
language = {english},
}
```

### Supported Tasks and Leaderboards

Grounding Visual Question Answering

### Language

English

## Dataset Structure

### Data Instances
A data instance contains a URL to the picture, information about the image size - width and height, information about the ground truth bounding box - left top and right bottom dots, and contains the question related to the picture.
```
{'image': https://toloka-cdn.azureedge.net/wsdmcup2023/000000000013.jpg,
'width': 640,
'height': 427,
'left': 129,
'top': 192,
'right': 155,
'bottom': 212,
'question': What does it use to breath?}
```
### Data Fields

* image: contains URL to the image
* width: value in pixels of image width
* height: value in pixels of image height
* left: the x coordinate in pixels to determine the left-top dot of the bounding box
* top: the y coordinate in pixels to determine the left-top dot of the bounding box
* right: the x coordinate in pixels to determine the right-bottom dot of the bounding box
* bottom: the y coordinate in pixels to determine the right-bottom dot of the bounding box
* question: a question related to the picture

### Data Splits
There are four splits in the data: train, train_sample, test_public, and test_private. 'train' split contains the full pull for model training.
The 'train-sample' split contains the part of the 'train' split. The 'test_public' split contains public data to test the model.
The 'test_private' split contains private data for the final model test.

### Source Data

The images were obtained from a CC BY-licensed subset of the Microsoft Common Objects in
Context dataset, [MS COCO](https://cocodataset.org/).

### Annotations

All data labeling was performed on the [Toloka crowdsourcing platform](https://toloka.ai/).

Only annotators who self-reported the knowledge of English had access to the annotation task.

### Citation Information

* Competition: https://toloka.ai/challenges/wsdm2023
* CodaLab: https://codalab.lisn.upsaclay.fr/competitions/7434
* Dataset: https://doi.org/10.5281/zenodo.7057740

原始信息汇总

数据集概述

名称: WSDMCup2023

语言: 英语

许可证: CC-BY-4.0

多语言性: 单语种

大小: 10K<n<100K

任务类别: 视觉问答

数据集结构

数据实例

每个数据实例包含以下字段：

• image: 图片的URL
• width: 图片宽度（像素）
• height: 图片高度（像素）
• left: 边界框左上角x坐标（像素）
• top: 边界框左上角y坐标（像素）
• right: 边界框右下角x坐标（像素）
• bottom: 边界框右下角y坐标（像素）
• question: 与图片相关的问题

数据字段

• image: 图片URL
• width: 图片宽度（像素）
• height: 图片高度（像素）
• left: 边界框左上角x坐标（像素）
• top: 边界框左上角y坐标（像素）
• right: 边界框右下角x坐标（像素）
• bottom: 边界框右下角y坐标（像素）
• question: 与图片相关的问题

数据分割

• train: 38,990个实例
• train_sample: 1,000个实例
• test_public: 1,705个实例
• test_private: 4,504个实例

数据来源

图片来源于CC BY-licensed的Microsoft Common Objects in Context (MS COCO) 数据集。

注释信息

所有数据标注在Toloka crowdsourcing平台完成。

引用信息

bibtex @inproceedings{TolokaWSDMCup2023, author = {Ustalov, Dmitry and Pavlichenko, Nikita and Koshelev, Sergey and Likhobaba, Daniil and Smirnova, Alisa}, title = {{Toloka Visual Question Answering Benchmark}}, year = {2023}, eprint = {2309.16511}, eprinttype = {arxiv}, eprintclass = {cs.CV}, language = {english}, }