five

CyberHarem/hibiki_bluearchive

收藏
Hugging Face2024-03-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/CyberHarem/hibiki_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 hibiki/猫塚ヒビキ/响 (Blue Archive) This is the dataset of hibiki/猫塚ヒビキ/响 (Blue Archive), containing 500 images and their tags. The core tags of this character are `long_hair, animal_ears, black_hair, dog_ears, dog_girl, halo, breasts, goggles_on_head, dog_tail, tail, yellow_halo, blue_eyes, eyewear_on_head, medium_breasts, ponytail`, 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 | 925.73 MiB | [Download](https://huggingface.co/datasets/CyberHarem/hibiki_bluearchive/resolve/main/dataset-raw.zip) | Waifuc-Raw | Raw data with meta information (min edge aligned to 1400 if larger). | | 1200 | 500 | 763.90 MiB | [Download](https://huggingface.co/datasets/CyberHarem/hibiki_bluearchive/resolve/main/dataset-1200.zip) | IMG+TXT | dataset with the shorter side not exceeding 1200 pixels. | | stage3-p480-1200 | 1411 | 1.60 GiB | [Download](https://huggingface.co/datasets/CyberHarem/hibiki_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/hibiki_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 | 6 | ![](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) | 1girl, bare_shoulders, blue_sky, blush, cleavage, confetti, crop_top, day, goggles, holding_pom_poms, looking_at_viewer, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, outdoors, pleated_skirt, solo, stomach, white_skirt, arm_up, large_breasts, open_mouth, armpits, thighs, cloud, sticker_on_face, sweat | | 1 | 5 | ![](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, bare_shoulders, cleavage, cowboy_shot, crop_top, goggles, holding_pom_poms, looking_at_viewer, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, pleated_skirt, solo, sticker_on_face, stomach, white_skirt, blush, open_mouth, simple_background | | 2 | 14 | ![](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, bare_shoulders, blush, crop_top, goggles, holding_pom_poms, midriff, millennium_cheerleader_outfit_(blue_archive), navel, official_alternate_costume, pleated_skirt, simple_background, solo, white_background, white_skirt, cleavage, looking_at_viewer, miniskirt, cowboy_shot, open_mouth, stomach, large_breasts, sticker_on_face | | 3 | 5 | ![](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, bare_shoulders, blush, crop_top, goggles, holding_pom_poms, looking_at_viewer, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, pleated_skirt, simple_background, sneakers, solo, white_background, white_footwear, white_skirt, cleavage, stomach, full_body, open_mouth | | 4 | 8 | ![](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, ass, bare_shoulders, blue_sky, blush, crop_top, goggles, holding_pom_poms, looking_at_viewer, looking_back, millennium_cheerleader_outfit_(blue_archive), miniskirt, official_alternate_costume, solo, thighs, white_panties, white_skirt, day, from_behind, outdoors, sticker_on_face, sweat, open_mouth, pleated_skirt, cloud, cowboy_shot, star_sticker, confetti, standing | | 5 | 5 | ![](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, bare_shoulders, blush, clothed_sex, crop_top, goggles, hetero, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, pleated_skirt, solo_focus, sweat, white_skirt, cleavage, cowgirl_position, girl_on_top, holding_pom_poms, open_mouth, sticker_on_face, stomach, vaginal, clothed_female_nude_male, looking_at_viewer, penis, spread_legs, large_breasts, mosaic_censoring, no_panties, pov, pussy | | 6 | 8 | ![](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, closed_mouth, fishnet_pantyhose, goggles, sitting, solo, blush, hair_ornament, looking_at_viewer, off_shoulder, black_footwear, boots, long_sleeves, simple_background, bare_shoulders, collar, dog_tags, purple_eyes, white_background, white_jacket, black_shorts, full_body, twintails | ### Table Version | # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | 1girl | bare_shoulders | blue_sky | blush | cleavage | confetti | crop_top | day | goggles | holding_pom_poms | looking_at_viewer | midriff | millennium_cheerleader_outfit_(blue_archive) | miniskirt | navel | official_alternate_costume | outdoors | pleated_skirt | solo | stomach | white_skirt | arm_up | large_breasts | open_mouth | armpits | thighs | cloud | sticker_on_face | sweat | cowboy_shot | simple_background | white_background | sneakers | white_footwear | full_body | ass | looking_back | white_panties | from_behind | star_sticker | standing | 1boy | clothed_sex | hetero | solo_focus | cowgirl_position | girl_on_top | vaginal | clothed_female_nude_male | penis | spread_legs | mosaic_censoring | no_panties | pov | pussy | closed_mouth | fishnet_pantyhose | sitting | hair_ornament | off_shoulder | black_footwear | boots | long_sleeves | collar | dog_tags | purple_eyes | white_jacket | black_shorts | twintails | |----:|----------:|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------|:-----------------|:-----------|:--------|:-----------|:-----------|:-----------|:------|:----------|:-------------------|:--------------------|:----------|:-----------------------------------------------|:------------|:--------|:-----------------------------|:-----------|:----------------|:-------|:----------|:--------------|:---------|:----------------|:-------------|:----------|:---------|:--------|:------------------|:--------|:--------------|:--------------------|:-------------------|:-----------|:-----------------|:------------|:------|:---------------|:----------------|:--------------|:---------------|:-----------|:-------|:--------------|:---------|:-------------|:-------------------|:--------------|:----------|:---------------------------|:--------|:--------------|:-------------------|:-------------|:------|:--------|:---------------|:--------------------|:----------|:----------------|:---------------|:-----------------|:--------|:---------------|:---------|:-----------|:--------------|:---------------|:---------------|:------------| | 0 | 6 | ![](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 | X | X | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1 | 5 | ![](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 | 14 | ![](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 | X | | X | X | | | | X | | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 3 | 5 | ![](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 | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 4 | 8 | ![](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 | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 5 | 5 | ![](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 | X | X | X | | | | | | | | | | | | | | | | 6 | 8 | ![](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 | X |
提供机构:
CyberHarem
原始信息汇总

数据集概述:hibiki/猫塚ヒビキ/响 (Blue Archive)

数据集基本信息

  • 许可证: MIT
  • 任务类别: 文本到图像
  • 标签: 艺术, 不适用于所有观众
  • 大小类别: 小于1K

数据集内容

  • 包含内容: 500张图像及其标签
  • 核心标签: long_hair, animal_ears, black_hair, dog_ears, dog_girl, halo, breasts, goggles_on_head, dog_tail, tail, yellow_halo, blue_eyes, eyewear_on_head, medium_breasts, ponytail

数据集下载

  • raw: 500张图像, 925.73 MiB

    • 下载链接: raw
    • 类型: Waifuc-Raw
    • 描述: 包含元信息的原始数据(最小边对齐到1400像素,如果更大)
  • 1200: 500张图像, 763.90 MiB

    • 下载链接: 1200
    • 类型: IMG+TXT
    • 描述: 短边不超过1200像素的数据集
  • stage3-p480-1200: 1411张图像, 1.60 GiB

    • 下载链接: stage3-p480-1200
    • 类型: IMG+TXT
    • 描述: 三阶段裁剪数据集,区域不小于480x480像素

数据集加载

  • 加载工具: waifuc

  • 加载代码示例: python import os import zipfile

    from huggingface_hub import hf_hub_download from waifuc.source import LocalSource

    下载原始档案文件

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

    提取文件到你的目录

    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])

数据集集群列表

  • 集群示例:
    • 集群0: 6个样本, 包含标签: 1girl, bare_shoulders, blue_sky, blush, cleavage, confetti, crop_top, day, goggles, holding_pom_poms, looking_at_viewer, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, outdoors, pleated_skirt, solo, stomach, white_skirt, arm_up, large_breasts, open_mouth, armpits, thighs, cloud, sticker_on_face, sweat
    • 集群1: 5个样本, 包含标签: 1girl, bare_shoulders, cleavage, cowboy_shot, crop_top, goggles, holding_pom_poms, looking_at_viewer, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, pleated_skirt, solo, sticker_on_face, stomach, white_skirt, blush, open_mouth, simple_background
    • 集群2: 14个样本, 包含标签: 1girl, bare_shoulders, blush, crop_top, goggles, holding_pom_poms, midriff, millennium_cheerleader_outfit_(blue_archive), navel, official_alternate_costume, pleated_skirt, simple_background, solo, white_background, white_skirt, cleavage, looking_at_viewer, miniskirt, cowboy_shot, open_mouth, stomach, large_breasts, sticker_on_face
    • 集群3: 5个样本, 包含标签: 1girl, bare_shoulders, blush, crop_top, goggles, holding_pom_poms, looking_at_viewer, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, pleated_skirt, simple_background, sneakers, solo, white_background, white_footwear, white_skirt, cleavage, stomach, full_body, open_mouth
    • 集群4: 8个样本, 包含标签: 1girl, ass, bare_shoulders, blue_sky, blush, crop_top, goggles, holding_pom_poms, looking_at_viewer, looking_back, millennium_cheerleader_outfit_(blue_archive), miniskirt, official_alternate_costume, solo, thighs, white_panties, white_skirt, day, from_behind, outdoors, sticker_on_face, sweat, open_mouth, pleated_skirt, cloud, cowboy_shot, star_sticker, confetti, standing
    • 集群5: 5个样本, 包含标签: 1boy, 1girl, bare_shoulders, blush, clothed_sex, crop_top, goggles, hetero, midriff, millennium_cheerleader_outfit_(blue_archive), miniskirt, navel, official_alternate_costume, pleated_skirt, solo_focus, sweat, white_skirt, cleavage, cowgirl_position, girl_on_top, holding_pom_poms, open_mouth, sticker_on_face, stomach, vaginal, clothed_female_nude_male, looking_at_viewer, penis, spread_legs, large_breasts, mosaic_censoring, no_panties, pov, pussy
    • 集群6: 8个样本, 包含标签: 1girl, closed_mouth, fishnet_pantyhose, goggles, sitting, solo, blush, hair_ornament, looking_at_viewer, off_shoulder, black_footwear, boots, long_sleeves, simple_background, bare_shoulders, collar, dog_tags, purple_eyes, white_background, white_jacket, black_shorts, full_body, twintails
搜集汇总
数据集介绍
main_image_url
构建方式
在图像生成与动漫文化交融的背景下,CyberHarem/hibiki_bluearchive 数据集应运而生,专注于收录《蔚蓝档案》中角色“响”的视觉素材。该数据集通过由 DeepGHS 团队开发的自动化爬取系统,从 Danbooru、Pixiv、Zerochan 等多个平台采集图像,最终精选出 500 张高质量图片及其对应的标签信息。数据集的核心标签如长发、兽耳、黑发、犬耳、光环等已在构建过程中被精简处理,以聚焦于角色的本质特征。数据以多种格式打包发布,包括保留元数据的原始格式、短边不超过 1200 像素的标准化版本,以及经三阶段裁剪、面积不低于 480×480 像素的高效训练格式,为不同应用场景提供了灵活选择。
使用方法
该数据集的使用方式兼顾了便捷性与可扩展性。对于偏好原生加载的用户,可通过 Waifuc 框架直接读取原始数据包,利用其提供的 API 便捷地访问图像、文件名与标签元数据。具体操作时,需先从 Hugging Face Hub 下载压缩包,解压至本地目录后,使用 LocalSource 进行迭代处理。标准化与裁剪版本则以图像加文本文件的格式提供,可直接用于大多数深度学习框架的常规训练流程,无需额外依赖。研究者亦可基于标签聚类结果,快速定位特定主题的子集进行针对性实验,例如筛选“啦啦队服”簇以训练风格条件生成模型。整个流程从下载到使用均保持了高度的透明性与可复现性。
背景与挑战
背景概述
该数据集由DeepGHS团队于近年创建,专注于《碧蓝档案》游戏角色hibiki(猫塚ヒビキ/响)的图像与标签收集。核心研究问题在于为文本到图像生成模型提供高质量、多视角的人物图像素材,以推动二次元角色生成领域的发展。数据集包含500张经过精细标注的图像,标签覆盖角色外貌特征(如长发、兽耳、光环)与场景细节,并通过聚类分析展示了不同服饰与姿态的分布。其影响力体现在为动漫风格图像生成任务提供了标准化数据基准,尤其适用于角色一致性生成与多模态模型训练。
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,文本到图像生成任务中角色身份保持一致性的难题,例如在多种服饰(如啦啦队服、日常装)与姿势下维持hibiki的核心特征(狗耳、黑发)不丢失;2)构建过程中,数据来源分散(danbooru、pixiv等多站点)导致标签噪声与图像质量参差不齐,需通过自动化爬虫与人工校验平衡效率与准确性;3)数据集规模较小(500张),难以覆盖角色所有可能姿态与场景,限制了模型泛化能力;4)部分内容涉及敏感标签(如成人主题),需谨慎处理以符合伦理与使用规范。
常用场景
经典使用场景
该数据集聚焦于《蔚蓝档案》中角色猫塚响(Hibiki)的视觉素材,收录了500张高质量图像及其对应的标签信息。在文本到图像生成领域,研究者常将其用于微调扩散模型或训练个性化角色生成器,借助其丰富的姿态、服饰与场景标签,可有效提升模型对特定动漫角色特征的捕捉能力与生成一致性。
解决学术问题
该数据集解决了动漫角色细粒度图像生成中标注数据稀缺与风格泛化性不足的难题。通过提供多源爬取、标签聚类与多尺度裁剪的标准化流程,它为条件图像生成、少样本学习及风格迁移等学术任务提供了可复现的基准资源,推动了虚拟角色个性化生成技术向精细化与可控化方向演进。
实际应用
在实际应用中,该数据集支撑了虚拟偶像内容创作、游戏角色二次开发与同人艺术自动化生产等场景。创作者可基于该数据集训练轻量化生成模型,快速产出符合角色核心特征(如兽耳、光环、啦啦队服)的衍生图像,显著降低了动漫IP衍生内容的人力与时间成本。
数据集最近研究
最新研究方向
在二次元角色生成与个性化图像合成领域,基于《蔚蓝档案》中角色“响”的高质量数据集正成为推动文本到图像模型精细调优的关键资源。该数据集包含500张经过多源爬取与标签精炼的图像,其核心特征如长发、兽耳、光环及多种服饰变体(如啦啦队服)被系统性地聚合,为角色一致性生成提供了结构化支撑。当前前沿研究聚焦于利用此类专有数据集进行概念注入与风格迁移,结合扩散模型实现角色在不同场景下的高保真渲染。同时,聚类分析揭示的服装与姿态分布(如官方变体与同人创作)正被用于探索模型对细微视觉差异的捕捉能力,这对提升虚拟角色在交互式生成、游戏资产自动化创作中的表现力具有显著意义,也推动了少样本学习在动漫风格生成中的实际应用。
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