CyberHarem/noelle_genshin

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

# Dataset of noelle/ノエル/诺艾尔 (Genshin Impact)

This is the dataset of noelle/ノエル/诺艾尔 (Genshin Impact), containing 500 images and their tags.

The core tags of this character are `grey_hair, green_eyes, breasts, braid, large_breasts, short_hair, wavy_hair, medium_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 | Images-others | Images-head |
|:-----------------|---------:|:---------|:----------------------------------------------------------------------------------------------------------------|:-----------|:---------------------------------------------------------------------|:----------------|:--------------|
| raw | 500 | 1.02 GiB | [Download](https://huggingface.co/datasets/CyberHarem/noelle_genshin/resolve/main/dataset-raw.zip) | Waifuc-Raw | Raw data with meta information (min edge aligned to 1400 if larger). | -- | -- |
| stage3-p480-1200 | 1324 | 1.74 GiB | [Download](https://huggingface.co/datasets/CyberHarem/noelle_genshin/resolve/main/dataset-stage3-p480-1200.zip) | IMG+TXT | 3-stage cropped dataset with the area not less than 480x480 pixels. | 856 | 468 |

### 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/noelle_genshin',
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 | 28 |  |  |  |  |  | 1girl, collarbone, hair_ribbon, looking_at_viewer, official_alternate_hairstyle, virtual_youtuber, white_sweater, black_choker, plaid_skirt, ribbed_sweater, blue_ribbon, cleavage, official_alternate_costume, solo, off-shoulder_sweater, blush, smile, closed_mouth, strap_between_breasts, simple_background, brown_skirt, french_braid, handbag, black_skirt, white_background, shoulder_bag, glasses |
| 1 | 6 |  |  |  |  |  | 1girl, cleavage, cow_ears, cow_horns, cow_print, cowbell, looking_at_viewer, neck_bell, official_alternate_hairstyle, solo, virtual_youtuber, blush, official_alternate_costume, white_sweater, bare_shoulders, black_skirt, blue_bow, blunt_bangs, cow_tail, long_sleeves, open_mouth, :d, choker, clothing_cutout, hair_bow, holding_bottle, milk_bottle, off-shoulder_sweater, plaid_skirt, tail_bow |
| 2 | 28 |  |  |  |  |  | collared_shirt, looking_at_viewer, school_uniform, white_shirt, 1girl, blue_bowtie, pleated_skirt, solo, virtual_youtuber, blush, official_alternate_hairstyle, plaid_skirt, long_sleeves, smile, closed_mouth, official_alternate_costume, sitting, x_hair_ornament, blue_skirt, indoors, open_clothes, shirt_tucked_in, striped_bow |
| 3 | 7 |  |  |  |  |  | 1girl, bare_shoulders, detached_sleeves, looking_at_viewer, solo, virtual_youtuber, official_alternate_costume, belt, blue_ribbon, blush, closed_mouth, hair_ribbon, holding, long_hair, official_alternate_hair_length, ponytail, smile, thigh_strap, black_shorts, cowboy_shot, see-through_cleavage, short_shorts, weapon, white_dress, hair_ornament, long_sleeves, outdoors, white_shirt |
| 4 | 14 |  |  |  |  |  | 1girl, cleavage, frills, official_alternate_costume, official_alternate_hair_length, solo, virtual_youtuber, beer_mug, black_bowtie, hair_flower, holding_cup, looking_at_viewer, very_long_hair, corset, dirndl, blush, open_mouth, waist_apron, white_apron, detached_collar, :d, detached_sleeves, short_sleeves, blunt_bangs, simple_background, skirt, white_background, white_flower, capelet, cow_ears, cow_horns, dress, puffy_sleeves, upper_body, upper_teeth_only |
| 5 | 8 |  |  |  |  |  | 1girl, black_bowtie, cleavage, hair_flower, looking_at_viewer, official_alternate_costume, official_alternate_hair_length, smile, solo, virtual_youtuber, blush, detached_sleeves, dirndl, short_sleeves, very_long_hair, frills, closed_mouth, corset, outdoors, blurry_background, detached_collar, dress, upper_body, waist_apron, white_apron, white_thighhighs |
| 6 | 16 |  |  |  |  |  | 1girl, looking_at_viewer, solo, virtual_youtuber, blush, cleavage_cutout, shoulder_armor, smile, black_gloves, fingerless_gloves, chest_belt, mole_on_breast, tiara, hair_between_eyes, upper_body, simple_background, brown_belt, closed_mouth, open_mouth, vambraces, white_background, pouch |
| 7 | 21 |  |  |  |  |  | cleavage, looking_at_viewer, virtual_youtuber, 1girl, solo, blush, eyewear_on_head, sunglasses, bikini_skirt, navel, choker, collarbone, open_mouth, outdoors, two-tone_bikini, arm_strap, day, mole_on_breast, thigh_strap, :d, blue_sky, cloud, front-tie_bikini_top, hair_between_eyes, water, beach, ocean |
| 8 | 10 |  |  |  |  |  | 1girl, hololive_idol_uniform, official_alternate_costume, solo, tiara, white_gloves, bare_shoulders, looking_at_viewer, sleeveless, smile, blush, idol_clothes, open_mouth, bowtie, navel, white_skirt, wrist_cuffs, hair_ribbon, miniskirt, belt, blue_bow, holding, midriff |
| 9 | 7 |  |  |  |  |  | 1girl, armored_dress, cleavage, gauntlets, hair_flower, looking_at_viewer, maid_headdress, medium_breasts, red_rose, smile, solo, apron, closed_mouth, red_ascot, shoulder_armor, blush, white_background, armored_boots, braided_bangs, simple_background |
| 10 | 9 |  |  |  |  |  | 1girl, beret, black_headwear, long_sleeves, looking_at_viewer, official_alternate_costume, solo, blush, open_mouth, :d, black_dress, employee_uniform, white_pantyhose, blunt_bangs, burger, disposable_cup, holding, white_apron, bow, detached_collar, drinking_straw, red_ascot, tray |
| 11 | 15 |  |  |  |  |  | 1girl, solo, virtual_youtuber, looking_at_viewer, obi, blue_kimono, hair_flower, official_alternate_hairstyle, print_kimono, smile, wide_sleeves, blush, alternate_costume, closed_mouth, floral_print, long_sleeves, outdoors, upper_body, white_flower, hair_between_eyes |

### Table Version

| # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | 1girl | collarbone | hair_ribbon | looking_at_viewer | official_alternate_hairstyle | virtual_youtuber | white_sweater | black_choker | plaid_skirt | ribbed_sweater | blue_ribbon | cleavage | official_alternate_costume | solo | off-shoulder_sweater | blush | smile | closed_mouth | strap_between_breasts | simple_background | brown_skirt | french_braid | handbag | black_skirt | white_background | shoulder_bag | glasses | cow_ears | cow_horns | cow_print | cowbell | neck_bell | bare_shoulders | blue_bow | blunt_bangs | cow_tail | long_sleeves | open_mouth | :d | choker | clothing_cutout | hair_bow | holding_bottle | milk_bottle | tail_bow | collared_shirt | school_uniform | white_shirt | blue_bowtie | pleated_skirt | sitting | x_hair_ornament | blue_skirt | indoors | open_clothes | shirt_tucked_in | striped_bow | detached_sleeves | belt | holding | long_hair | official_alternate_hair_length | ponytail | thigh_strap | black_shorts | cowboy_shot | see-through_cleavage | short_shorts | weapon | white_dress | hair_ornament | outdoors | frills | beer_mug | black_bowtie | hair_flower | holding_cup | very_long_hair | corset | dirndl | waist_apron | white_apron | detached_collar | short_sleeves | skirt | white_flower | capelet | dress | puffy_sleeves | upper_body | upper_teeth_only | blurry_background | white_thighhighs | cleavage_cutout | shoulder_armor | black_gloves | fingerless_gloves | chest_belt | mole_on_breast | tiara | hair_between_eyes | brown_belt | vambraces | pouch | eyewear_on_head | sunglasses | bikini_skirt | navel | two-tone_bikini | arm_strap | day | blue_sky | cloud | front-tie_bikini_top | water | beach | ocean | hololive_idol_uniform | white_gloves | sleeveless | idol_clothes | bowtie | white_skirt | wrist_cuffs | miniskirt | midriff | armored_dress | gauntlets | maid_headdress | medium_breasts | red_rose | apron | red_ascot | armored_boots | braided_bangs | beret | black_headwear | black_dress | employee_uniform | white_pantyhose | burger | disposable_cup | bow | drinking_straw | tray | obi | blue_kimono | print_kimono | wide_sleeves | alternate_costume | floral_print |
|----:|----------:|:---------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------|:--------|:-------------|:--------------|:--------------------|:-------------------------------|:-------------------|:----------------|:---------------|:--------------|:-----------------|:--------------|:-----------|:-----------------------------|:-------|:-----------------------|:--------|:--------|:---------------|:------------------------|:--------------------|:--------------|:---------------|:----------|:--------------|:-------------------|:---------------|:----------|:-----------|:------------|:------------|:----------|:------------|:-----------------|:-----------|:--------------|:-----------|:---------------|:-------------|:-----|:---------|:------------------|:-----------|:-----------------|:--------------|:-----------|:-----------------|:-----------------|:--------------|:--------------|:----------------|:----------|:------------------|:-------------|:----------|:---------------|:------------------|:--------------|:-------------------|:-------|:----------|:------------|:---------------------------------|:-----------|:--------------|:---------------|:--------------|:-----------------------|:---------------|:---------|:--------------|:----------------|:-----------|:---------|:-----------|:---------------|:--------------|:--------------|:-----------------|:---------|:---------|:--------------|:--------------|:------------------|:----------------|:--------|:---------------|:----------|:--------|:----------------|:-------------|:-------------------|:--------------------|:-------------------|:------------------|:-----------------|:---------------|:--------------------|:-------------|:-----------------|:--------|:--------------------|:-------------|:------------|:--------|:------------------|:-------------|:---------------|:--------|:------------------|:------------|:------|:-----------|:--------|:-----------------------|:--------|:--------|:--------|:------------------------|:---------------|:-------------|:---------------|:---------|:--------------|:--------------|:------------|:----------|:----------------|:------------|:-----------------|:-----------------|:-----------|:--------|:------------|:----------------|:----------------|:--------|:-----------------|:--------------|:-------------------|:------------------|:---------|:-----------------|:------|:-----------------|:-------|:------|:--------------|:---------------|:---------------|:--------------------|:---------------|
| 0 | 28 |  |  |  |  |  | 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 | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 1 | 6 |  |  |  |  |  | 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 | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 2 | 28 |  |  |  |  |  | X | | | X | X | X | | | X | | | | X | X | | X | X | X | | | | | | | | | | | | | | | | | | | X | | | | | | | | | X | X | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 3 | 7 |  |  |  |  |  | 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 | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 4 | 14 |  |  |  |  |  | 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 | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 5 | 8 |  |  |  |  |  | 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 | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 6 | 16 |  |  |  |  |  | X | | | X | | X | | | | | | | | X | | X | X | X | | X | | | | | X | | | | | | | | | | | | | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | X | | | | X | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 7 | 21 |  |  |  |  |  | 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 | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| 8 | 10 |  |  |  |  |  | X | | X | X | | | | | | | | | X | X | | X | X | | | | | | | | | | | | | | | | X | X | | | | X | | | | | | | | | | | | | | | | | | | | | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | X | | | | | | | | X | | | | | | | | | | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | |
| 9 | 7 |  |  |  |  |  | X | | | X | | | | | | | | X | | X | | X | X | X | | X | | | | | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | X | | | | | | | | | | | | | | | | | | | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | |
| 10 | 9 |  |  |  |  |  | X | | | X | | | | | | | | | X | X | | X | | | | | | | | | | | | | | | | | | | X | | X | X | X | | | | | | | | | | | | | | | | | | | | | X | | | | | | | | | | | | | | | | | | | | | | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | X | | | X | X | X | X | X | X | X | X | X | X | | | | | | |
| 11 | 15 |  |  |  |  |  | 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

基本信息

• 名称: noelle/ノエル/诺艾尔 (Genshin Impact)
• 许可证: MIT
• 任务类别: text-to-image
• 标签: art, not-for-all-audiences
• 大小类别: n<1K

数据集内容

• 包含内容: 500张图像及其标签
• 核心标签: grey_hair, green_eyes, breasts, braid, large_breasts, short_hair, wavy_hair, medium_hair

数据来源

• 图像来源: 多个网站（如danbooru, pixiv, zerochan等）
• 采集系统: 由DeepGHS Team提供

数据集结构

包名	图像数量	大小	下载链接	类型	描述
raw	500	1.02 GiB	下载链接	Waifuc-Raw	包含元信息的原始数据（最小边对齐到1400像素）
stage3-p480-1200	1324	1.74 GiB	下载链接	IMG+TXT	三阶段裁剪数据集，区域不小于480x480像素

使用方法

• 加载原始数据集: 提供了一个使用waifuc加载原始数据集的Python示例代码。

数据集示例

• 集群列表: 提供了多个集群的示例图像及其详细标签，展示了数据集中的多样性和标签的丰富性。

结论

该数据集适用于text-to-image任务，特别关注于Genshin Impact中的角色noelle/ノエル/诺艾尔，提供了丰富的图像和详细标签，适合用于相关研究和应用。

搜集汇总

数据集介绍

构建方式

在数字内容创作领域，高质量的角色图像数据集对于文本到图像生成模型的训练至关重要。本数据集聚焦于《原神》中备受喜爱的角色诺艾尔，通过自动化爬取系统从Danbooru、Pixiv、Zerochan等多个知名图站采集原始图像，并由DeepGHS团队提供技术支持。数据集共收录500张图像及其对应的标签信息，核心标签涵盖灰发、绿瞳、辫子等角色特征，并经过精心剪裁以突出关键视觉元素。此外，数据集提供了两种封装形式：原始压缩包包含带元数据的全尺寸图像，而进阶版本则经过三级裁剪处理，生成面积不小于480×480像素的图像-文本配对数据，其中全身图像856张、头部图像468张，为不同训练需求提供了灵活选择。

特点

该数据集呈现出鲜明的结构化与精细化特点。首先，所有图像均附带详尽标签，不仅包含角色基础属性，还通过聚类分析揭示了11种不同的服装与场景变体，如校园制服、泳装、和服等，为多风格生成任务提供了丰富素材。其次，数据集设计了多阶段处理流程，原始数据保留完整元信息，而裁剪版本则专注于主体区域，有效降低了背景噪声对模型训练的干扰。最后，标签系统与Waifuc框架深度兼容，使得数据加载与处理流程高度自动化，研究者可便捷地利用标签进行条件生成或风格迁移实验。

使用方法

使用本数据集时，推荐采用Waifuc工具链进行高效加载。研究者可通过Hugging Face Hub下载原始压缩包，利用Python的zipfile模块解压至本地目录，随后借助Waifuc的LocalSource接口读取图像与标签数据，实现迭代式访问。对于进阶用户，可直接使用三级裁剪后的图像-文本配对数据，该格式已适配主流扩散模型的训练范式。此外，数据集提供的聚类结果可作为先验知识，辅助设计针对特定服装或场景的微调策略，例如通过标签过滤提取泳装子集进行风格迁移训练。整体而言，该数据集兼顾了易用性与扩展性，能够满足从基础研究到应用开发的多层次需求。

背景与挑战

背景概述

在文本到图像生成领域，高质量、标注精细的角色数据集对于训练可控的生成模型至关重要。CyberHarem团队于近期构建了名为noelle_genshin的数据集，专注于收录热门游戏《原神》中角色诺艾尔（Noelle）的视觉素材。该数据集由DeepGHS团队主导开发，通过自动化爬虫系统从Danbooru、Pixiv、Zerochan等多个知名艺术平台采集了500张图像及其对应的标签信息。核心研究问题聚焦于如何为二次元角色提供结构化的多模态训练数据，以支撑下游的个性化图像生成任务。该数据集不仅包含了原始图像与元数据，还提供了经过多阶段裁剪的版本，有效提升了数据利用率，对动漫风格的角色生成研究产生了积极的推动作用。

当前挑战

该数据集所面临的挑战首先体现在领域问题的复杂性上：文本到图像生成任务要求模型能够精准理解并还原角色的核心外观特征（如灰发、绿眼、辫子等），但二次元角色在不同画师笔下存在显著的风格差异，导致模型在泛化与保真度之间难以平衡。其次，在数据集构建过程中，自动化爬取面临图像来源多样、版权归属模糊以及标签噪声高等问题。此外，原始图像分辨率参差不齐，需经过复杂的多阶段裁剪与过滤流程才能生成符合训练标准的高质量样本，而最终仅收录500张图像，数据规模有限，可能影响模型对角色丰富姿态与服饰变体的学习能力。

常用场景

经典使用场景

在文本到图像生成领域，CyberHarem/noelle_genshin数据集为基于扩散模型的角色定制化生成提供了高质量的微调素材。该数据集收录了《原神》中诺艾尔这一角色的500张精细图像及对应标签，涵盖多种官方与同人衍生服饰、姿态与场景，适用于训练模型精准捕捉角色核心特征（如灰发、绿瞳、发辫等），并支持在Stable Diffusion等架构上通过DreamBooth或LoRA方法实现角色一致性生成。

解决学术问题

该数据集有效解决了动漫角色图像生成中常见的身份特征保持难题，为学者研究多模态对齐与细粒度概念学习提供了标准化基准。通过结构化标签与多阶段裁剪处理，它助力探究条件生成模型在有限样本下的泛化能力，并推动了关于视觉特征解耦、风格迁移与属性组合等学术议题的深入探索。

衍生相关工作

该数据集衍生出多项经典工作，包括基于Waifuc框架的自动化爬取与标注管线、多阶段图像裁剪与标签聚类方法，以及用于角色服装分类的簇分析结果。这些工作为后续角色数据集构建（如CyberHarem系列）奠定了技术范式，并启发了关于动漫图像质量筛选、标签层级优化与少样本概念学习的后续研究。