five

CyberHarem/chinatsu_bluearchive

收藏
Hugging Face2024-03-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/CyberHarem/chinatsu_bluearchive
下载链接
链接失效反馈
官方服务:
资源简介:
--- license: mit task_categories: - text-to-image tags: - art - not-for-all-audiences size_categories: - n<1K --- # Dataset of chinatsu/火宮チナツ/千夏 (Blue Archive) This is the dataset of chinatsu/火宮チナツ/千夏 (Blue Archive), containing 265 images and their tags. The core tags of this character are `long_hair, pointy_ears, breasts, brown_hair, halo, large_breasts, glasses, hairband, yellow_eyes, black-framed_eyewear, brown_eyes, twintails`, 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 | 265 | 462.13 MiB | [Download](https://huggingface.co/datasets/CyberHarem/chinatsu_bluearchive/resolve/main/dataset-raw.zip) | Waifuc-Raw | Raw data with meta information (min edge aligned to 1400 if larger). | | 1200 | 265 | 385.88 MiB | [Download](https://huggingface.co/datasets/CyberHarem/chinatsu_bluearchive/resolve/main/dataset-1200.zip) | IMG+TXT | dataset with the shorter side not exceeding 1200 pixels. | | stage3-p480-1200 | 667 | 789.46 MiB | [Download](https://huggingface.co/datasets/CyberHarem/chinatsu_bluearchive/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/chinatsu_bluearchive', 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 | 30 | ![](samples/0/clu0-sample0.png) | ![](samples/0/clu0-sample1.png) | ![](samples/0/clu0-sample2.png) | ![](samples/0/clu0-sample3.png) | ![](samples/0/clu0-sample4.png) | looking_at_viewer, 1girl, blush, solo, collarbone, onsen, no_eyewear, water, wet, closed_mouth, smile, cleavage, covering_privates, snow, red_halo, sitting, nude_cover, steam, hair_over_shoulder, white_towel | | 1 | 12 | ![](samples/1/clu1-sample0.png) | ![](samples/1/clu1-sample1.png) | ![](samples/1/clu1-sample2.png) | ![](samples/1/clu1-sample3.png) | ![](samples/1/clu1-sample4.png) | 1girl, long_sleeves, onsen, solo, white_kimono, looking_at_viewer, wide_sleeves, blush, closed_mouth, sash, sitting, official_alternate_costume, rock, water, bath_yukata, black_hairband, cleavage, smile, wooden_bucket, collarbone, thighs | | 2 | 7 | ![](samples/2/clu2-sample0.png) | ![](samples/2/clu2-sample1.png) | ![](samples/2/clu2-sample2.png) | ![](samples/2/clu2-sample3.png) | ![](samples/2/clu2-sample4.png) | 1girl, blush, looking_at_viewer, solo, white_kimono, cleavage, closed_mouth, official_alternate_costume, smile, collarbone, long_sleeves, bath_yukata, off_shoulder, wide_sleeves, black_hairband, bow | | 3 | 6 | ![](samples/3/clu3-sample0.png) | ![](samples/3/clu3-sample1.png) | ![](samples/3/clu3-sample2.png) | ![](samples/3/clu3-sample3.png) | ![](samples/3/clu3-sample4.png) | 1girl, blush, off_shoulder, simple_background, solo, white_background, white_kimono, bare_shoulders, black_hairband, cleavage, collarbone, looking_at_viewer, official_alternate_costume, sweat, yukata, closed_mouth, no_eyewear, sitting, smile | | 4 | 13 | ![](samples/4/clu4-sample0.png) | ![](samples/4/clu4-sample1.png) | ![](samples/4/clu4-sample2.png) | ![](samples/4/clu4-sample3.png) | ![](samples/4/clu4-sample4.png) | 1girl, blush, nipples, 1boy, hetero, solo_focus, penis, navel, pussy, on_back, spread_legs, open_mouth, sex, vaginal, completely_nude, looking_at_viewer, mosaic_censoring, yukata, bar_censor, collarbone, missionary, onsen, pov, pubic_hair, smile, sweat | | 5 | 13 | ![](samples/5/clu5-sample0.png) | ![](samples/5/clu5-sample1.png) | ![](samples/5/clu5-sample2.png) | ![](samples/5/clu5-sample3.png) | ![](samples/5/clu5-sample4.png) | 1boy, 1girl, hetero, low_twintails, nipples, solo_focus, white_shirt, bow, bra_lift, collarbone, open_shirt, red_pantyhose, simple_background, torn_pantyhose, white_background, white_bra, armband, navel, no_eyewear, pov, blush, penis, uncensored, girl_on_top, cowgirl_position, open_mouth, pussy, looking_at_viewer, sex, vaginal, imminent_penetration | | 6 | 16 | ![](samples/6/clu6-sample0.png) | ![](samples/6/clu6-sample1.png) | ![](samples/6/clu6-sample2.png) | ![](samples/6/clu6-sample3.png) | ![](samples/6/clu6-sample4.png) | 1girl, black_skirt, collared_shirt, long_sleeves, looking_at_viewer, red_gloves, solo, white_shirt, armband, pleated_skirt, red_pantyhose, blush, red_ribbon, neck_ribbon, low_twintails, closed_mouth, light_brown_hair, simple_background, sitting, red_halo, white_background, hair_between_eyes | | 7 | 7 | ![](samples/7/clu7-sample0.png) | ![](samples/7/clu7-sample1.png) | ![](samples/7/clu7-sample2.png) | ![](samples/7/clu7-sample3.png) | ![](samples/7/clu7-sample4.png) | 1girl, armband, bag, holding_gun, pleated_skirt, red_gloves, red_pantyhose, ribbon, solo, white_shirt, black_skirt, looking_at_viewer, boots, long_sleeves, simple_background, white_background, collared_shirt, full_body, low_twintails, closed_mouth | ### Table Version | # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | looking_at_viewer | 1girl | blush | solo | collarbone | onsen | no_eyewear | water | wet | closed_mouth | smile | cleavage | covering_privates | snow | red_halo | sitting | nude_cover | steam | hair_over_shoulder | white_towel | long_sleeves | white_kimono | wide_sleeves | sash | official_alternate_costume | rock | bath_yukata | black_hairband | wooden_bucket | thighs | off_shoulder | bow | simple_background | white_background | bare_shoulders | sweat | yukata | nipples | 1boy | hetero | solo_focus | penis | navel | pussy | on_back | spread_legs | open_mouth | sex | vaginal | completely_nude | mosaic_censoring | bar_censor | missionary | pov | pubic_hair | low_twintails | white_shirt | bra_lift | open_shirt | red_pantyhose | torn_pantyhose | white_bra | armband | uncensored | girl_on_top | cowgirl_position | imminent_penetration | black_skirt | collared_shirt | red_gloves | pleated_skirt | red_ribbon | neck_ribbon | light_brown_hair | hair_between_eyes | bag | holding_gun | ribbon | boots | full_body | |----:|----------:|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------|:--------|:--------|:-------|:-------------|:--------|:-------------|:--------|:------|:---------------|:--------|:-----------|:--------------------|:-------|:-----------|:----------|:-------------|:--------|:---------------------|:--------------|:---------------|:---------------|:---------------|:-------|:-----------------------------|:-------|:--------------|:-----------------|:----------------|:---------|:---------------|:------|:--------------------|:-------------------|:-----------------|:--------|:---------|:----------|:-------|:---------|:-------------|:--------|:--------|:--------|:----------|:--------------|:-------------|:------|:----------|:------------------|:-------------------|:-------------|:-------------|:------|:-------------|:----------------|:--------------|:-----------|:-------------|:----------------|:-----------------|:------------|:----------|:-------------|:--------------|:-------------------|:-----------------------|:--------------|:-----------------|:-------------|:----------------|:-------------|:--------------|:-------------------|:--------------------|:------|:--------------|:---------|:--------|:------------| | 0 | 30 | ![](samples/0/clu0-sample0.png) | ![](samples/0/clu0-sample1.png) | ![](samples/0/clu0-sample2.png) | ![](samples/0/clu0-sample3.png) | ![](samples/0/clu0-sample4.png) | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1 | 12 | ![](samples/1/clu1-sample0.png) | ![](samples/1/clu1-sample1.png) | ![](samples/1/clu1-sample2.png) | ![](samples/1/clu1-sample3.png) | ![](samples/1/clu1-sample4.png) | X | X | X | X | X | X | | X | | X | X | X | | | | X | | | | | X | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 2 | 7 | ![](samples/2/clu2-sample0.png) | ![](samples/2/clu2-sample1.png) | ![](samples/2/clu2-sample2.png) | ![](samples/2/clu2-sample3.png) | ![](samples/2/clu2-sample4.png) | X | X | X | X | X | | | | | X | X | X | | | | | | | | | X | X | X | | X | | X | X | | | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 3 | 6 | ![](samples/3/clu3-sample0.png) | ![](samples/3/clu3-sample1.png) | ![](samples/3/clu3-sample2.png) | ![](samples/3/clu3-sample3.png) | ![](samples/3/clu3-sample4.png) | X | X | X | X | X | | X | | | X | X | X | | | | X | | | | | | X | | | X | | | X | | | X | | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 4 | 13 | ![](samples/4/clu4-sample0.png) | ![](samples/4/clu4-sample1.png) | ![](samples/4/clu4-sample2.png) | ![](samples/4/clu4-sample3.png) | ![](samples/4/clu4-sample4.png) | 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 | 13 | ![](samples/5/clu5-sample0.png) | ![](samples/5/clu5-sample1.png) | ![](samples/5/clu5-sample2.png) | ![](samples/5/clu5-sample3.png) | ![](samples/5/clu5-sample4.png) | 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 | | | | | | | | | | | | | | | 6 | 16 | ![](samples/6/clu6-sample0.png) | ![](samples/6/clu6-sample1.png) | ![](samples/6/clu6-sample2.png) | ![](samples/6/clu6-sample3.png) | ![](samples/6/clu6-sample4.png) | X | X | X | X | | | | | | X | | | | | X | X | | | | | X | | | | | | | | | | | | X | X | | | | | | | | | | | | | | | | | | | | | | X | X | | | X | | | X | | | | | X | X | X | X | X | X | X | X | | | | | | | 7 | 7 | ![](samples/7/clu7-sample0.png) | ![](samples/7/clu7-sample1.png) | ![](samples/7/clu7-sample2.png) | ![](samples/7/clu7-sample3.png) | ![](samples/7/clu7-sample4.png) | X | X | | X | | | | | | X | | | | | | | | | | | X | | | | | | | | | | | | X | X | | | | | | | | | | | | | | | | | | | | | | X | X | | | X | | | X | | | | | X | X | X | X | | | | | X | X | X | X | X |
提供机构:
CyberHarem
原始信息汇总

数据集概述:chinatsu/火宮チナツ/千夏 (Blue Archive)

基本信息

  • 数据集名称:chinatsu/火宮チナツ/千夏 (Blue Archive)
  • 包含内容:265张图像及其标签
  • 核心标签:long_hair, pointy_ears, breasts, brown_hair, halo, large_breasts, glasses, hairband, yellow_eyes, black-framed_eyewear, brown_eyes, twintails
  • 许可证:MIT
  • 任务类别:text-to-image
  • 标签:art, not-for-all-audiences
  • 大小类别:n<1K

数据集包详情

名称 图像数量 大小 下载链接 类型 描述
raw 265 462.13 MiB 下载 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)。
1200 265 385.88 MiB 下载 IMG+TXT 短边不超过1200像素的数据集。
stage3-p480-1200 667 789.46 MiB 下载 IMG+TXT 三阶段裁剪数据集,区域不小于480x480像素。

数据集使用示例

数据集支持通过waifuc加载。示例代码如下:

python import os import zipfile from huggingface_hub import hf_hub_download from waifuc.source import LocalSource

下载原始存档文件

zip_file = hf_hub_download( repo_id=CyberHarem/chinatsu_bluearchive, repo_type=dataset, filename=dataset-raw.zip, )

提取文件到目录

dataset_dir = dataset_dir os.makedirs(dataset_dir, exist_ok=True) with zipfile.ZipFile(zip_file, r) as zf: zf.extractall(dataset_dir)

使用waifuc加载数据集

source = LocalSource(dataset_dir) for item in source: print(item.image, item.meta[filename], item.meta[tags])

数据集标签集群

数据集提供了标签集群结果,有助于进一步的数据挖掘。以下是部分集群示例:

集群示例

# 样本数 图像示例 标签示例
0 30 looking_at_viewer, 1girl, blush, solo, collarbone, onsen, no_eyewear, water, wet, closed_mouth, smile, cleavage, covering_privates, snow, red_halo, sitting, nude_cover, steam, hair_over_shoulder, white_towel
1 12 1girl, long_sleeves, onsen, solo, white_kimono, looking_at_viewer, wide_sleeves, blush, closed_mouth, sash, sitting, official_alternate_costume, rock, water, bath_yukata, black_hairband, cleavage, smile, wooden_bucket, collarbone, thighs
2 7 1girl, blush, looking_at_viewer, solo, white_kimono, cleavage, closed_mouth, official_alternate_costume, smile, collarbone, long_sleeves, bath_yukata, off_shoulder, wide_sleeves, black_hairband, bow
3 6 1girl, blush, off_shoulder, simple_background, solo, white_background, white_kimono, bare_shoulders, black_hairband, cleavage, collarbone, looking_at_viewer, official_alternate_costume, sweat, yukata, closed_mouth, no_eyewear, sitting, smile
4 13 1girl, blush, nipples, 1boy, hetero, solo_focus, penis, navel, pussy, on_back, spread_legs, open_mouth, sex, vaginal, completely_nude, looking_at_viewer, mosaic_censoring, yukata, bar_censor, collarbone, missionary, onsen, pov, pubic_hair, smile, sweat
5 13 1boy, 1girl, hetero, low_twintails, nipples, solo_focus, white_shirt, bow, bra_lift, collarbone, open_shirt, red_pantyhose, simple_background, torn_pantyhose, white_background, white_bra, armband, navel, no_eyewear, pov, blush, penis, uncensored, girl_on_top, cowgirl_position, open_mouth, pussy, looking_at_viewer, sex, vaginal, imminent_penetration
6 16 1girl, black_skirt, collared_shirt, long_sleeves, looking_at_viewer, red_gloves, solo, white_shirt, armband, pleated_skirt, red_pantyhose, blush, red_ribbon, neck_ribbon, low_twintails, closed_mouth, light_brown_hair, simple_background, sitting, red_halo, white_background, hair_between_eyes
7 7 1girl, armband, bag, holding_gun, pleated_skirt, red_gloves, red_pantyhose, ribbon, solo, white_shirt, black_skirt, looking_at_viewer, boots, long_sleeves, simple_background, white_background, collared_shirt, full_body, low_twintails, closed_mouth
搜集汇总
数据集介绍
main_image_url
构建方式
本数据集聚焦于游戏《蔚蓝档案》中的角色火宮チナツ(千夏),通过自动化爬取系统从Danbooru、Pixiv、Zerochan等多个图像平台搜集了265张高质量图片及对应的标签信息。数据集的构建依托于DeepGHS团队的自动化技术,确保了图像来源的广泛性与多样性。原始数据经过预处理,包含元数据信息,并提供了多种规格的压缩包,包括原始分辨率版本、短边不超过1200像素的版本,以及经过三阶段裁剪、确保最小区域不低于480x480像素的增强版本,以适配不同应用场景。
使用方法
数据集支持通过waifuc库进行原始数据的加载,用户可下载ZIP压缩包并利用LocalSource接口读取图像及其关联的标签信息。对于需要直接使用图像-文本对进行模型训练的场景,推荐采用1200像素或三阶段裁剪版本,这些版本已标准化处理,可直接用于文本到图像生成任务。用户可通过HuggingFace Hub的API获取数据,并参照提供的代码示例快速集成至自己的工作流中,实现高效的数据调用与实验复现。
背景与挑战
背景概述
该数据集由DeepGHS团队于近期创建,聚焦于《蔚蓝档案》中角色火宮千夏(Chinatsu)的视觉表征学习。作为文本到图像生成领域的一项基础资源,数据集收录了265张来自Danbooru、Pixiv等多平台的二次元图像,并附有结构化标签。其核心研究问题在于为特定虚拟角色构建高质量的图像-文本配对数据,以支撑扩散模型等生成式架构的微调与风格迁移。数据集提供了raw、1200及stage3-crop三种规格的版本,并通过waifuc框架支持灵活加载,在动漫风格生成与角色定制领域具有显著影响力。
当前挑战
当前数据集面临的主要挑战包括:1)领域问题层面,需解决动漫角色生成中细节保真度与风格一致性的矛盾,尤其是角色核心特征(如发型、光环、眼镜)在多样化场景下的鲁棒表征;2)构建过程中,图像爬取自多源平台,面临版权合规、元数据噪声及标签语义冲突(如部分图像含敏感内容)的挑战;3)数据集规模较小(265张),难以覆盖角色全部姿态与着装变体,可能限制生成模型的泛化能力;4)标签聚类结果显示内容异构性显著(如温泉、战斗、亲密场景混杂),需进一步清洗与领域适配。
常用场景
经典使用场景
在动漫角色生成与个性化创作领域,CyberHarem/chinatsu_bluearchive数据集为文本到图像(text-to-image)任务提供了高质量、精标注的视觉素材。该数据集聚焦于《蔚蓝档案》中的角色千夏(火宮チナツ),包含265张经过精细裁剪与标签清洗的图像,其核心特征如长耳、光环、眼镜等被系统化提取,为基于扩散模型或生成对抗网络的细粒度角色生成研究奠定了数据基础。研究者可借助此数据集训练模型,实现从文本描述到特定角色形象的精准映射,从而推动动漫风格图像合成技术的边界拓展。
解决学术问题
该数据集有效回应了动漫图像生成领域中长期存在的角色一致性保持与多模态对齐难题。传统数据集往往缺乏对单一角色多姿态、多场景的覆盖,导致生成结果在服装、发型等细节上出现混淆。CyberHarem/chinatsu_bluearchive通过聚类分析将图像按角色服饰与场景(如温泉、制服)分组,并附有结构化标签体系,使学术研究能够聚焦于如何利用有限样本(n<1K)实现高保真角色再现。其意义在于为小样本学习、领域自适应及跨模态检索提供了标准化测试基准,推动了生成模型在亚文化场景中的学术验证。
实际应用
在实际应用中,该数据集可服务于游戏开发、虚拟偶像制作与个性化内容创作等产业环节。例如,游戏美术团队可基于此数据集微调Stable Diffusion等模型,快速生成角色在多种情境下的概念图或宣传素材,大幅降低手工绘制的成本与周期。此外,结合waifuc等工具,数据集支持流水线式的图像获取与预处理,便于集成到自动化内容生产系统中,为UGC社区提供定制化角色生成服务,从而在娱乐与商业场景中释放创意生产力。
数据集最近研究
最新研究方向
在二次元角色生成与文本到图像(Text-to-Image)技术快速迭代的背景下,CyberHarem/chinatsu_bluearchive数据集聚焦于《蔚蓝档案》中角色“千夏”的高质量多模态数据构建,代表了游戏角色定制化图像生成的前沿探索。该数据集包含265张精心筛选的图像及其标签,覆盖从原始高分辨率到多阶段裁剪的多种规格,并创新性地提供了基于Waifuc框架的原始数据加载接口,极大便利了角色特定扩散模型的微调与风格迁移研究。尤其值得关注的是,通过标签聚类分析,数据集揭示了角色在不同服饰(如浴衣、校服)与场景(如温泉、战斗)下的视觉特征分布,这为解耦角色身份与外观属性的可控生成技术提供了关键支撑。当前,围绕该数据集的关联热点包括利用LoRA等轻量级微调方法实现角色一致性保持,以及通过多标签条件控制提升生成图像的语义对齐精度,其影响已延伸至游戏美术资产自动化生产与虚拟角色个性化创作领域,为二次元文化生态中的AI辅助内容生成树立了新的数据基准。
以上内容由遇见数据集搜集并总结生成
二维码
社区交流群
二维码
科研交流群
商业服务