cats_vs_dogs

# Dataset Card for Cats Vs. Dogs

## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
- [Dataset Summary](#dataset-summary)
- [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
- [Languages](#languages)
- [Dataset Structure](#dataset-structure)
- [Data Instances](#data-instances)
- [Data Fields](#data-fields)
- [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
- [Curation Rationale](#curation-rationale)
- [Source Data](#source-data)
- [Annotations](#annotations)
- [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
- [Social Impact of Dataset](#social-impact-of-dataset)
- [Discussion of Biases](#discussion-of-biases)
- [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
- [Dataset Curators](#dataset-curators)
- [Licensing Information](#licensing-information)
- [Citation Information](#citation-information)
- [Contributions](#contributions)

## Dataset Description

- **Homepage:** [Cats vs Dogs Dataset](https://www.microsoft.com/en-us/download/details.aspx?id=54765)
- **Repository:**
- **Paper:** [Asirra: A CAPTCHA that Exploits Interest-Aligned Manual Image Categorization](https://www.microsoft.com/en-us/research/wp-content/uploads/2007/10/CCS2007.pdf)
- **Leaderboard:** [Dogs vs. Cats](https://www.kaggle.com/competitions/dogs-vs-cats)
- **Point of Contact:**

### Dataset Summary

A large set of images of cats and dogs. There are 1738 corrupted images that are dropped. This dataset is part of a now-closed Kaggle competition and represents a subset of the so-called Asirra dataset.

From the competition page:

> The Asirra data set
>
> Web services are often protected with a challenge that's supposed to be easy for people to solve, but difficult for computers. Such a challenge is often called a [CAPTCHA](http://www.captcha.net/) (Completely Automated Public Turing test to tell Computers and Humans Apart) or HIP (Human Interactive Proof). HIPs are used for many purposes, such as to reduce email and blog spam and prevent brute-force attacks on web site passwords.
>
> Asirra (Animal Species Image Recognition for Restricting Access) is a HIP that works by asking users to identify photographs of cats and dogs. This task is difficult for computers, but studies have shown that people can accomplish it quickly and accurately. Many even think it's fun! Here is an example of the Asirra interface:
>
> Asirra is unique because of its partnership with [Petfinder.com](https://www.petfinder.com/), the world's largest site devoted to finding homes for homeless pets. They've provided Microsoft Research with over three million images of cats and dogs, manually classified by people at thousands of animal shelters across the United States. Kaggle is fortunate to offer a subset of this data for fun and research.

### Supported Tasks and Leaderboards

- `image-classification`: The goal of this task is to classify a given image as either containing a cat or a dog. The leaderboard is available [here](https://paperswithcode.com/sota/image-classification-on-cats-vs-dogs).

### Languages

English.

## Dataset Structure

### Data Instances

A sample from the training set is provided below:

```
{
'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=500x375 at 0x29CEAD71780>,
'labels': 0
}
```

### Data Fields

The data instances have the following fields:

- `image`: A `PIL.Image.Image` object containing the image. Note that when accessing the image column: `dataset[0]["image"]` the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the `"image"` column, *i.e.* `dataset[0]["image"]` should **always** be preferred over `dataset["image"][0]`.
- `labels`: an `int` classification label.

Class Label Mappings:

```
{
"cat": 0,
"dog": 1,
}
```

### Data Splits

| | train |
|---------------|------:|
| # of examples | 23410 |

## Dataset Creation

### Curation Rationale

This subset was to built to test whether computer vision algorithms can beat the Asirra CAPTCHA:

From the competition page:

> Image recognition attacks
>
> While random guessing is the easiest form of attack, various forms of image recognition can allow an attacker to make guesses that are better than random. There is enormous diversity in the photo database (a wide variety of backgrounds, angles, poses, lighting, etc.), making accurate automatic classification difficult. In an informal poll conducted many years ago, computer vision experts posited that a classifier with better than 60% accuracy would be difficult without a major advance in the state of the art. For reference, a 60% classifier improves the guessing probability of a 12-image HIP from 1/4096 to 1/459.

### Source Data

#### Initial Data Collection and Normalization

This dataset is a subset of the Asirra dataset.

From the competition page:

> Asirra is unique because of its partnership with Petfinder.com, the world's largest site devoted to finding homes for homeless pets. They've provided Microsoft Research with over three million images of cats and dogs, manually classified by people at thousands of animal shelters across the United States.

#### Who are the source language producers?

The users of [Petfinder.com](https://www.petfinder.com/).

### Annotations

#### Annotation process

The images were annotated by selecting a pet category on [Petfinder.com](https://www.petfinder.com/).

#### Who are the annotators?

The users of [Petfinder.com](https://www.petfinder.com/).

### Personal and Sensitive Information

[More Information Needed]

## Considerations for Using the Data

### Social Impact of Dataset

[More Information Needed]

### Discussion of Biases

From the paper:

> Unlike many image-based CAPTCHAs which are abstract or subjective, Asirra’s challenges are concrete, inoffensive (cute, by some accounts), require no specialized or culturally biased knowledge, and have definite ground truth. This
makes Asirra less frustrating for humans. Some beta-testers found it fun. The four-year-old child of one asked several times to “play the cat and dog game again.”

### Other Known Limitations

[More Information Needed]

## Additional Information

### Dataset Curators

[More Information Needed]

### Licensing Information

[More Information Needed]

### Citation Information

```bibtex
@Inproceedings (Conference){asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization,
author = {Elson, Jeremy and Douceur, John (JD) and Howell, Jon and Saul, Jared},
title = {Asirra: A CAPTCHA that Exploits Interest-Aligned Manual Image Categorization},
booktitle = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},
year = {2007},
month = {October},
publisher = {Association for Computing Machinery, Inc.},
url = {https://www.microsoft.com/en-us/research/publication/asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization/},
edition = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},
}
```

### Contributions

Thanks to [@nateraw](https://github.com/nateraw) for adding this dataset.

# 猫与狗数据集卡片

## 目录
- [目录](#目录)
- [数据集描述](#数据集描述)
- [数据集概述](#数据集概述)
- [支持任务与排行榜](#支持任务与排行榜)
- [语言](#语言)
- [数据集结构](#数据集结构)
- [数据样本](#数据样本)
- [数据字段](#数据字段)
- [数据划分](#数据划分)
- [数据集构建](#数据集构建)
- [构建逻辑](#构建逻辑)
- [源数据](#源数据)
- [标注信息](#标注信息)
- [个人与敏感信息](#个人与敏感信息)
- [数据集使用注意事项](#数据集使用注意事项)
- [数据集的社会影响](#数据集的社会影响)
- [偏差讨论](#偏差讨论)
- [其他已知局限性](#其他已知局限性)
- [附加信息](#附加信息)
- [数据集维护者](#数据集维护者)
- [许可信息](#许可信息)
- [引用信息](#引用信息)
- [贡献致谢](#贡献致谢)

## 数据集描述
- **主页:** [猫 vs 狗数据集](https://www.microsoft.com/en-us/download/details.aspx?id=54765)
- **代码仓库:**
- **相关论文:** [Asirra: 一种基于兴趣对齐人工图像分类的CAPTCHA（全自动区分计算机和人类的图灵测试）](https://www.microsoft.com/en-us/research/wp-content/uploads/2007/10/CCS2007.pdf)
- **排行榜:** [狗 vs 猫](https://www.kaggle.com/competitions/dogs-vs-cats)
- **联系人:**

### 数据集概述
本数据集包含大量猫与狗的图像，其中1738张损坏图像已被剔除。该数据集是已停办的Kaggle竞赛的一部分，为所谓的Asirra数据集的子集。

摘自竞赛页面：

> ### Asirra数据集
> 网络服务通常会设置一类挑战，要求人类可轻松完成，但计算机难以破解，这类挑战常被称为**CAPTCHA（全自动区分计算机和人类的图灵测试，Completely Automated Public Turing test to tell Computers and Humans Apart）**或**HIP（人机交互验证，Human Interactive Proof）**。HIP可用于多种场景，例如减少邮件与博客垃圾信息，以及防止针对网站密码的暴力破解攻击。
>
> Asirra（动物物种识别访问限制，Animal Species Image Recognition for Restricting Access）是一类HIP，其通过要求用户识别猫与狗的照片来完成验证。该任务对计算机而言难度较高，但研究表明人类可以快速且准确地完成，甚至不少用户认为该过程颇具乐趣。以下为Asirra界面示例：
>
> Asirra的独特之处在于其与全球最大的流浪宠物领养平台Petfinder.com达成合作。该平台向微软研究院提供了超过300万张由美国数千家动物收容所工作人员人工分类的猫与狗图像。Kaggle有幸提供该数据集的子集用于学术研究与趣味实验。

### 支持任务与排行榜
- `图像分类（image-classification）`: 该任务的目标是将给定图像分类为猫或狗。排行榜可参见[此处](https://paperswithcode.com/sota/image-classification-on-cats-vs-dogs)。

### 语言
英语。

## 数据集结构
### 数据样本
训练集的一个样本示例如下：

{
'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=500x375 at 0x29CEAD71780>,
'labels': 0
}

### 数据字段
数据样本包含以下字段：
- `image`: 包含图像的`PIL.Image.Image`对象。请注意，当访问图像列时：`dataset[0]["image"]`会自动对图像文件进行解码。解码大量图像文件可能会耗费大量时间，因此建议优先通过样本索引再指定图像列，即始终优先使用`dataset[0]["image"]`而非`dataset["image"][0]`。
- `labels`: 整数类型的分类标签。

类别标签映射：

{
"cat": 0,
"dog": 1,
}

### 数据划分
| | 训练集 |
|---------------|------:|
| 样本数量 | 23410 |

## 数据集构建
### 构建逻辑
构建该子集的目的是验证计算机视觉算法是否可以破解Asirra CAPTCHA：
> 摘自竞赛页面：
>
> > 图像识别攻击
> > 虽然随机猜测是最简单的攻击方式，但各类图像识别技术可以让攻击者获得比随机猜测更高的准确率。该图像数据库包含极高的多样性（多样的背景、拍摄角度、姿态、光照条件等），这使得全自动精准分类极具挑战。多年前的一项非正式调查显示，计算机视觉专家认为，若不取得当前技术的重大突破，想要构建准确率超过60%的分类器将极为困难。作为参考，准确率60%的分类器可以将12图像HIP任务的猜测概率从1/4096提升至1/459。

### 源数据
#### 初始数据收集与标准化
本数据集是Asirra数据集的子集。
> 摘自竞赛页面：
>
> > Asirra的独特之处在于其与全球最大的流浪宠物领养平台Petfinder.com达成合作。该平台向微软研究院提供了超过300万张由美国数千家动物收容所工作人员人工分类的猫与狗图像。

#### 源语言生产者是谁？
Petfinder.com的用户。

### 标注信息
#### 标注流程
图像通过在Petfinder.com上选择宠物类别完成标注。
#### 标注者是谁？
Petfinder.com的用户。

### 个人与敏感信息
[需补充更多信息]

## 数据集使用注意事项
### 数据集的社会影响
[需补充更多信息]

### 偏差讨论
> 摘自相关论文：
>
> > 与许多基于抽象或主观图像的CAPTCHA不同，Asirra的验证任务具体且无害（部分用户认为其画面可爱），无需专门或受文化局限的知识，且拥有明确的真实标签。这使得人类用户的验证体验更为流畅，部分测试用户甚至觉得该过程充满乐趣。一名测试者的四岁孩子甚至多次要求“再玩一次猫狗识别游戏”。

### 其他已知局限性
[需补充更多信息]

## 附加信息
### 数据集维护者
[需补充更多信息]

### 许可信息
[需补充更多信息]

### 引用信息
bibtex
@Inproceedings (Conference){asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization,
author = {Elson, Jeremy and Douceur, John (JD) and Howell, Jon and Saul, Jared},
title = {Asirra: A CAPTCHA that Exploits Interest-Aligned Manual Image Categorization},
booktitle = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},
year = {2007},
month = {October},
publisher = {Association for Computing Machinery, Inc.},
url = {https://www.microsoft.com/en-us/research/publication/asirra-a-captcha-that-exploits-interest-aligned-manual-image-categorization/},
edition = {Proceedings of 14th ACM Conference on Computer and Communications Security (CCS)},
}

### 贡献致谢
感谢[@nateraw](https://github.com/nateraw) 添加本数据集。