CyberHarem/fuuro_pokemon

---
license: mit
task_categories:
- text-to-image
tags:
- art
- not-for-all-audiences
size_categories:
- n<1K
---

# Dataset of fuuro/フウロ (Pokémon)

This is the dataset of fuuro/フウロ (Pokémon), containing 500 images and their tags.

The core tags of this character are `red_hair, hair_ornament, breasts, blue_eyes, sidelocks, large_breasts, bangs, long_hair`, which are pruned in this dataset.

Images are crawled from many sites (e.g. danbooru, pixiv, zerochan ...), the auto-crawling system is powered by [DeepGHS Team](https://github.com/deepghs)([huggingface organization](https://huggingface.co/deepghs)).

## List of Packages

| Name | Images | Size | Download | Type | Description |
|:-----------------|---------:|:-----------|:---------------------------------------------------------------------------------------------------------------|:-----------|:---------------------------------------------------------------------|
| raw | 500 | 556.55 MiB | [Download](https://huggingface.co/datasets/CyberHarem/fuuro_pokemon/resolve/main/dataset-raw.zip) | Waifuc-Raw | Raw data with meta information (min edge aligned to 1400 if larger). |
| 800 | 500 | 324.82 MiB | [Download](https://huggingface.co/datasets/CyberHarem/fuuro_pokemon/resolve/main/dataset-800.zip) | IMG+TXT | dataset with the shorter side not exceeding 800 pixels. |
| stage3-p480-800 | 1248 | 700.58 MiB | [Download](https://huggingface.co/datasets/CyberHarem/fuuro_pokemon/resolve/main/dataset-stage3-p480-800.zip) | IMG+TXT | 3-stage cropped dataset with the area not less than 480x480 pixels. |
| 1200 | 500 | 495.26 MiB | [Download](https://huggingface.co/datasets/CyberHarem/fuuro_pokemon/resolve/main/dataset-1200.zip) | IMG+TXT | dataset with the shorter side not exceeding 1200 pixels. |
| stage3-p480-1200 | 1248 | 972.10 MiB | [Download](https://huggingface.co/datasets/CyberHarem/fuuro_pokemon/resolve/main/dataset-stage3-p480-1200.zip) | IMG+TXT | 3-stage cropped dataset with the area not less than 480x480 pixels. |

### Load Raw Dataset with Waifuc

We provide raw dataset (including tagged images) for [waifuc](https://deepghs.github.io/waifuc/main/tutorials/installation/index.html) loading. If you need this, just run the following code

```python
import os
import zipfile

from huggingface_hub import hf_hub_download
from waifuc.source import LocalSource

# download raw archive file
zip_file = hf_hub_download(
repo_id='CyberHarem/fuuro_pokemon',
repo_type='dataset',
filename='dataset-raw.zip',
)

# extract files to your directory
dataset_dir = 'dataset_dir'
os.makedirs(dataset_dir, exist_ok=True)
with zipfile.ZipFile(zip_file, 'r') as zf:
zf.extractall(dataset_dir)

# load the dataset with waifuc
source = LocalSource(dataset_dir)
for item in source:
print(item.image, item.meta['filename'], item.meta['tags'])
```

## List of Clusters

List of tag clustering result, maybe some outfits can be mined here.

### Raw Text Version

| # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | Tags |
|----:|----------:|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| 0 | 11 |  |  |  |  |  | 1girl, gloves, midriff, navel, blue_footwear, boots, crop_top, smile, solo, open_mouth, blue_shorts, short_shorts, simple_background, white_background |
| 1 | 15 |  |  |  |  |  | 1girl, :d, blue_gloves, blue_jacket, blue_shorts, cropped_jacket, looking_at_viewer, midriff, navel, one_side_up, open_mouth, short_hair_with_long_locks, upper_teeth_only, crop_top, short_shorts, solo, turtleneck, tongue, blush, eyelashes, hair_between_eyes, sky, boots, thigh_pouch, arm_up, cloud |
| 2 | 5 |  |  |  |  |  | 1girl, blush, smile, solo, gloves, open_mouth, looking_at_viewer |
| 3 | 5 |  |  |  |  |  | 1girl, navel, nipples, nude, one_side_up, collarbone, eyelashes, open_mouth, pussy, shiny_skin, short_hair_with_long_locks, solo, tongue, :d, blush, looking_at_viewer, mosaic_censoring, hand_up, hetero, upper_teeth_only |
| 4 | 24 |  |  |  |  |  | 1boy, 1girl, hetero, blush, solo_focus, nipples, penis, open_mouth, nude, sex, smile, navel, pussy, vaginal, looking_at_viewer, sweat, cum, mosaic_censoring, pov, spread_legs |
| 5 | 5 |  |  |  |  |  | 1girl, christmas, gloves, looking_at_viewer, official_alternate_costume, open_mouth, red_dress, santa_costume, smile, red_footwear, blush, long_sleeves, one_eye_closed, pokemon_(creature), short_hair_with_long_locks, solo, ;d, black_belt, feathers, gift_bag, holding, sack, santa_boots, strap_between_breasts, white_background |
| 6 | 5 |  |  |  |  |  | fake_animal_ears, playboy_bunny, rabbit_ears, 1girl, bowtie, detached_collar, wrist_cuffs, cleavage, open_mouth, pantyhose, smile, white_background, 2girls, alternate_costume, blue_leotard, dark_skin, rabbit_tail, salute, short_hair_with_long_locks, simple_background, solo_focus, strapless_leotard |

### Table Version

| # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | 1girl | gloves | midriff | navel | blue_footwear | boots | crop_top | smile | solo | open_mouth | blue_shorts | short_shorts | simple_background | white_background | :d | blue_gloves | blue_jacket | cropped_jacket | looking_at_viewer | one_side_up | short_hair_with_long_locks | upper_teeth_only | turtleneck | tongue | blush | eyelashes | hair_between_eyes | sky | thigh_pouch | arm_up | cloud | nipples | nude | collarbone | pussy | shiny_skin | mosaic_censoring | hand_up | hetero | 1boy | solo_focus | penis | sex | vaginal | sweat | cum | pov | spread_legs | christmas | official_alternate_costume | red_dress | santa_costume | red_footwear | long_sleeves | one_eye_closed | pokemon_(creature) | ;d | black_belt | feathers | gift_bag | holding | sack | santa_boots | strap_between_breasts | fake_animal_ears | playboy_bunny | rabbit_ears | bowtie | detached_collar | wrist_cuffs | cleavage | pantyhose | 2girls | alternate_costume | blue_leotard | dark_skin | rabbit_tail | salute | strapless_leotard |
|----:|----------:|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------|:---------|:----------|:--------|:----------------|:--------|:-----------|:--------|:-------|:-------------|:--------------|:---------------|:--------------------|:-------------------|:-----|:--------------|:--------------|:-----------------|:--------------------|:--------------|:-----------------------------|:-------------------|:-------------|:---------|:--------|:------------|:--------------------|:------|:--------------|:---------|:--------|:----------|:-------|:-------------|:--------|:-------------|:-------------------|:----------|:---------|:-------|:-------------|:--------|:------|:----------|:--------|:------|:------|:--------------|:------------|:-----------------------------|:------------|:----------------|:---------------|:---------------|:-----------------|:---------------------|:-----|:-------------|:-----------|:-----------|:----------|:-------|:--------------|:------------------------|:-------------------|:----------------|:--------------|:---------|:------------------|:--------------|:-----------|:------------|:---------|:--------------------|:---------------|:------------|:--------------|:---------|:--------------------|
| 0 | 11 |  |  |  |  |  | X | X | X | X | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 1 | 15 |  |  |  |  |  | X | | X | X | | X | X | | X | X | X | X | | | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 2 | 5 |  |  |  |  |  | X | X | | | | | | X | X | X | | | | | | | | | X | | | | | | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 3 | 5 |  |  |  |  |  | X | | | X | | | | | X | X | | | | | X | | | | X | X | X | X | | X | X | X | | | | | | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 4 | 24 |  |  |  |  |  | X | | | X | | | | X | | X | | | | | | | | | X | | | | | | X | | | | | | | X | X | | X | | X | | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 5 | 5 |  |  |  |  |  | X | X | | | | | | X | X | X | | | | X | | | | | X | | X | | | | X | | | | | | | | | | | | | | | | | | | | | | | | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | |
| 6 | 5 |  |  |  |  |  | X | | | | | | | X | | X | | | X | X | | | | | | | X | | | | | | | | | | | | | | | | | | | | X | | | | | | | | | | | | | | | | | | | | | | | | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X |

原始信息汇总

数据集概述

数据集信息

• 名称: Dataset of fuuro/フウロ (Pokémon)
• 描述: 包含500张图像及其标签的数据集。
• 核心标签: red_hair, hair_ornament, breasts, blue_eyes, sidelocks, large_breasts, bangs, long_hair
• 来源: 从多个网站（如danbooru, pixiv, zerochan等）爬取。
• 许可证: MIT
• 任务类别: text-to-image
• 标签: art, not-for-all-audiences
• 大小类别: n<1K

数据集包列表

名称	图像数量	大小	类型	描述
raw	500	556.55 MiB	Waifuc-Raw	包含元信息的原始数据（最小边对齐到1400像素，如果更大）。
800	500	324.82 MiB	IMG+TXT	短边不超过800像素的数据集。
stage3-p480-800	1248	700.58 MiB	IMG+TXT	3阶段裁剪数据集，区域不小于480x480像素。
1200	500	495.26 MiB	IMG+TXT	短边不超过1200像素的数据集。
stage3-p480-1200	1248	972.10 MiB	IMG+TXT	3阶段裁剪数据集，区域不小于480x480像素。

标签聚类结果

原始文本版本

#	样本数量	图像示例	标签
0	11		1girl, gloves, midriff, navel, blue_footwear, boots, crop_top, smile, solo, open_mouth, blue_shorts, short_shorts, simple_background, white_background
1	15		1girl, :d, blue_gloves, blue_jacket, blue_shorts, cropped_jacket, looking_at_viewer, midriff, navel, one_side_up, open_mouth, short_hair_with_long_locks, upper_teeth_only, crop_top, short_shorts, solo, turtleneck, tongue, blush, eyelashes, hair_between_eyes, sky, boots, thigh_pouch, arm_up, cloud
2	5		1girl, blush, smile, solo, gloves, open_mouth, looking_at_viewer
3	5		1girl, navel, nipples, nude, one_side_up, collarbone, eyelashes, open_mouth, pussy, shiny_skin, short_hair_with_long_locks, solo, tongue, :d, blush, looking_at_viewer, mosaic_censoring, hand_up, hetero, upper_teeth_only
4	24		1boy, 1girl, hetero, blush, solo_focus, nipples, penis, open_mouth, nude, sex, smile, navel, pussy, vaginal, looking_at_viewer, sweat, cum, mosaic_censoring, pov, spread_legs
5	5		1girl, christmas, gloves, looking_at_viewer, official_alternate_costume, open_mouth, red_dress, santa_costume, smile, red_footwear, blush, long_sleeves, one_eye_closed, pokemon_(creature), short_hair_with_long_locks, solo, ;d, black_belt, feathers, gift_bag, holding, sack, santa_boots, strap_between_breasts, white_background
6	5		fake_animal_ears, playboy_bunny, rabbit_ears, 1girl, bowtie, detached_collar, wrist_cuffs, cleavage, open_mouth, pantyhose, smile, white_background, 2girls, alternate_costume, blue_leotard, dark_skin, rabbit_tail, salute, short_hair_with_long_locks, simple_background, solo_focus, strapless_leotard

搜集汇总

数据集介绍

构建方式

在宝可梦这一经典动漫IP的视觉素材积累中，CyberHarem/fuuro_pokemon数据集专注于角色fuuro（フウロ）的图像与标签收集。该数据集通过自动化爬取系统从Danbooru、Pixiv、Zerochan等多个站点获取图像，系统由DeepGHS团队提供技术支撑。数据集包含500张原始图像及其对应的标签信息，核心标签如红发、发饰、胸部、蓝眼、侧发、巨乳、刘海、长发等已被精简处理。构建过程中，图像经过多尺度处理，衍生出多种规格版本，包括短边不超过800像素和1200像素的常规版本，以及经过三阶段裁剪、面积不小于480×480像素的增强版本，以适配不同训练需求。

特点

该数据集的一个显著特点在于其层次化的结构设计，提供了raw、800、1200及stage3-p480-800、stage3-p480-1200共五种打包形式，分别适用于原始元数据保留、标准训练与高精度裁剪场景。数据集的标签聚类功能尤为突出，通过将图像按视觉相似性与标签共现模式分组，展现了fuuro在不同服饰（如圣诞装、兔女郎装）和情境（如日常、亲密互动）下的多样化表现。每个聚类均附带示例图像与标签列表，便于用户快速识别特定主题或构图模式，为风格迁移、角色定制等文本到图像生成任务提供了结构化参考。

使用方法

使用该数据集时，用户可通过Hugging Face Hub下载对应的压缩包。对于原始数据，推荐采用waifuc库进行加载：首先利用huggingface_hub下载dataset-raw.zip，解压至本地目录后，通过LocalSource接口即可逐项获取图像及其元数据，包括文件名与标签。对于800、1200等规格版本，解压后直接获得图像与文本配对文件，适用于主流训练框架。聚类信息以表格形式提供，用户可依据聚类编号筛选特定风格的图像子集，从而高效地构建针对性的训练样本，提升模型对角色特定属性的生成质量。

背景与挑战

背景概述

在生成式人工智能浪潮中，文本到图像模型的发展对高质量、细粒度标注的图像数据集提出了迫切需求。CyberHarem/fuuro_pokemon数据集正是在此背景下，由DeepGHS团队于近期构建并发布，旨在服务于宝可梦角色“风露”的专属图像生成任务。该数据集汇集了500张来自Danbooru、Pixiv、Zerochan等平台的同人图像，并配备了精细的标签体系，核心标签涵盖红发、蓝眼、发饰等角色特征。其影响力体现在为二次元垂直领域的可控图像生成提供了标准化数据基础，推动了特定角色风格迁移与个性化生成的研究。

当前挑战

该数据集所面临的挑战主要体现在两个层面。在领域问题层面，如何从海量、风格迥异的同人图像中精准提取并统一角色核心视觉特征（如发型、服饰细节），以支撑稳定、高保真的文本到图像生成，是亟待解决的核心难题。在构建过程层面，数据采集依赖多源自动爬取，这带来了图像质量参差不齐、元数据格式不一以及版权归属模糊等挑战；此外，对包含成人内容的图像进行有效筛选与分级标注，确保数据集的合规使用，同样是构建过程中的关键难点。

常用场景

经典使用场景

在文本到图像生成领域，CyberHarem/fuuro_pokemon数据集为角色一致性生成提供了精良的视觉素材。该数据集收录了宝可梦角色フウロ的500张高质量图像及其标注标签，涵盖多种姿态、服饰与场景变体。研究者可借助此数据集训练扩散模型或生成对抗网络，使其精准捕捉该角色的核心视觉特征，如红发、蓝瞳与标志性发饰，从而在零样本或少样本条件下生成风格统一且细节丰富的角色图像。

实际应用

在实际应用中，该数据集赋能了游戏开发、虚拟偶像运营与二次元内容创作等领域的自动化生产流程。例如，创作者可基于此数据集微调模型，快速生成符合角色设定的宣传图、表情包或剧情插画，显著降低人工绘制成本。此外，其标签化结构也便于与推荐系统或交互式生成工具集成，实现用户驱动的个性化角色视觉输出。

衍生相关工作

围绕该数据集衍生了多项创新性工作。一方面，基于其聚类标签的研究催生了面向动漫角色的多模态语义理解方法，如利用标签层次结构改进图像-文本对齐模型。另一方面，数据集提供的三阶段裁剪版本（stage3-p480-800）为局部特征增强生成提供了基准，启发了注意力机制导向的角色部件生成网络。此外，其与waifuc工具的深度整合也推动了自动化数据流水线的标准化建设。