five

CyberHarem/miyu_bluearchive

收藏
Hugging Face2024-03-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/CyberHarem/miyu_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 miyu/霞沢ミユ/美游 (Blue Archive) This is the dataset of miyu/霞沢ミユ/美游 (Blue Archive), containing 500 images and their tags. The core tags of this character are `black_hair, long_hair, red_eyes, halo, leaf_on_head, ahoge, animal_ears, rabbit_ears, fake_animal_ears`, 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 | 989.53 MiB | [Download](https://huggingface.co/datasets/CyberHarem/miyu_bluearchive/resolve/main/dataset-raw.zip) | Waifuc-Raw | Raw data with meta information (min edge aligned to 1400 if larger). | | 1200 | 500 | 807.41 MiB | [Download](https://huggingface.co/datasets/CyberHarem/miyu_bluearchive/resolve/main/dataset-1200.zip) | IMG+TXT | dataset with the shorter side not exceeding 1200 pixels. | | stage3-p480-1200 | 1395 | 1.65 GiB | [Download](https://huggingface.co/datasets/CyberHarem/miyu_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/miyu_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 | 5 | ![](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, blue_sky, cloud, day, looking_at_viewer, navel, official_alternate_costume, outdoors, print_bikini, small_breasts, stomach, straw_hat, twin_braids, white_bikini, blush, closed_mouth, leaf_print, solo, ocean, seagull, sitting, white_pupils, beach, grey_halo, hat_on_back, twintails | | 1 | 22 | ![](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, leaf_print, official_alternate_costume, print_bikini, solo, twin_braids, white_bikini, navel, blush, looking_at_viewer, simple_background, white_background, grey_halo, flat_chest, straw_hat, white_pupils, closed_mouth, hat_on_back, stomach | | 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, looking_at_viewer, nipples, solo, blush, leaf, small_breasts, completely_nude, navel, uncensored, open_mouth, simple_background, white_background, cleft_of_venus, cum_in_pussy, cumdrip, stomach | | 3 | 10 | ![](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, black_gloves, blue_serafuku, blue_shirt, blue_skirt, blush, long_sleeves, white_pantyhose, white_sailor_collar, closed_mouth, green_neckerchief, leaf, looking_at_viewer, solo, pleated_skirt, simple_background, sitting, knee_pads, white_background | | 4 | 24 | ![](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, blue_serafuku, blue_skirt, long_sleeves, solo, white_pantyhose, black_gloves, blue_shirt, green_neckerchief, leaf, white_footwear, white_sailor_collar, looking_at_viewer, pleated_skirt, sneakers, bolt_action, knee_pads, sitting, closed_mouth, single_knee_pad, blush, simple_background, white_background, blunt_bangs, bright_pupils, full_body, holding_gun, trash_can | | 5 | 15 | ![](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) | 1girl, blue_shirt, blue_skirt, feet, long_sleeves, no_shoes, soles, solo, toes, white_pantyhose, black_gloves, blue_serafuku, blush, leaf, looking_at_viewer, panties_under_pantyhose, pleated_skirt, green_neckerchief, white_panties, white_sailor_collar, open_mouth, knee_pads, trash_can, simple_background, single_knee_pad, sitting, white_background, ass, blunt_bangs, rifle | | 6 | 9 | ![](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, alternate_costume, playboy_bunny, solo, strapless_leotard, bare_shoulders, detached_collar, looking_at_viewer, blush, rabbit_tail, simple_background, small_breasts, white_background, closed_mouth, white_pantyhose, wrist_cuffs, blue_leotard, bowtie, leaf, ass, fake_tail, no_shoes | ### Table Version | # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | 1girl | blue_sky | cloud | day | looking_at_viewer | navel | official_alternate_costume | outdoors | print_bikini | small_breasts | stomach | straw_hat | twin_braids | white_bikini | blush | closed_mouth | leaf_print | solo | ocean | seagull | sitting | white_pupils | beach | grey_halo | hat_on_back | twintails | simple_background | white_background | flat_chest | nipples | leaf | completely_nude | uncensored | open_mouth | cleft_of_venus | cum_in_pussy | cumdrip | black_gloves | blue_serafuku | blue_shirt | blue_skirt | long_sleeves | white_pantyhose | white_sailor_collar | green_neckerchief | pleated_skirt | knee_pads | white_footwear | sneakers | bolt_action | single_knee_pad | blunt_bangs | bright_pupils | full_body | holding_gun | trash_can | feet | no_shoes | soles | toes | panties_under_pantyhose | white_panties | ass | rifle | alternate_costume | playboy_bunny | strapless_leotard | bare_shoulders | detached_collar | rabbit_tail | wrist_cuffs | blue_leotard | bowtie | fake_tail | |----:|----------:|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------|:-----------|:--------|:------|:--------------------|:--------|:-----------------------------|:-----------|:---------------|:----------------|:----------|:------------|:--------------|:---------------|:--------|:---------------|:-------------|:-------|:--------|:----------|:----------|:---------------|:--------|:------------|:--------------|:------------|:--------------------|:-------------------|:-------------|:----------|:-------|:------------------|:-------------|:-------------|:-----------------|:---------------|:----------|:---------------|:----------------|:-------------|:-------------|:---------------|:------------------|:----------------------|:--------------------|:----------------|:------------|:-----------------|:-----------|:--------------|:------------------|:--------------|:----------------|:------------|:--------------|:------------|:-------|:-----------|:--------|:-------|:--------------------------|:----------------|:------|:--------|:--------------------|:----------------|:--------------------|:-----------------|:------------------|:--------------|:--------------|:---------------|:---------|:------------| | 0 | 5 | ![](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 | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1 | 22 | ![](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 | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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 | X | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 3 | 10 | ![](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 | 24 | ![](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 | | | | | | | | | | | | | | | | | | | | 5 | 15 | ![](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 | | | | | | | | | | | | 6 | 9 | ![](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 |
提供机构:
CyberHarem
原始信息汇总

数据集概述

数据集名称

Dataset of miyu/霞沢ミユ/美游 (Blue Archive)

数据集描述

该数据集包含500张图片及其标签,主要描述角色miyu/霞沢ミユ/美游(来自游戏Blue Archive)。

核心标签

  • black_hair
  • long_hair
  • red_eyes
  • halo
  • leaf_on_head
  • ahoge
  • animal_ears
  • rabbit_ears
  • fake_animal_ears

数据来源

图片从多个网站(如danbooru, pixiv, zerochan等)爬取,爬虫系统由DeepGHS Team开发。

数据集包列表

名称 图片数量 大小 类型 描述
raw 500 989.53 MiB Waifuc-Raw 包含元信息的原始数据(最小边对齐到1400像素)
1200 500 807.41 MiB IMG+TXT 短边不超过1200像素的数据集
stage3-p480-1200 1395 1.65 GiB IMG+TXT 3阶段裁剪数据集,区域不小于480x480像素

数据加载

提供原始数据集(包括标记的图片)供waifuc加载。

标签聚类结果

原始文本版本

# 样本数 图片1 图片2 图片3 图片4 图片5 标签
0 5 [图片] [图片] [图片] [图片] [图片] 1girl, blue_sky, cloud, day, looking_at_viewer, navel, official_alternate_costume, outdoors, print_bikini, small_breasts, stomach, straw_hat, twin_braids, white_bikini, blush, closed_mouth, leaf_print, solo, ocean, seagull, sitting, white_pupils, beach, grey_halo, hat_on_back, twintails
1 22 [图片] [图片] [图片] [图片] [图片] 1girl, leaf_print, official_alternate_costume, print_bikini, solo, twin_braids, white_bikini, navel, blush, looking_at_viewer, simple_background, white_background, grey_halo, flat_chest, straw_hat, white_pupils, closed_mouth, hat_on_back, stomach
2 7 [图片] [图片] [图片] [图片] [图片] 1girl, looking_at_viewer, nipples, solo, blush, leaf, small_breasts, completely_nude, navel, uncensored, open_mouth, simple_background, white_background, cleft_of_venus, cum_in_pussy, cumdrip, stomach
3 10 [图片] [图片] [图片] [图片] [图片] 1girl, black_gloves, blue_serafuku, blue_shirt, blue_skirt, blush, long_sleeves, white_pantyhose, white_sailor_collar, closed_mouth, green_neckerchief, leaf, looking_at_viewer, solo, pleated_skirt, simple_background, sitting, knee_pads, white_background
4 24 [图片] [图片] [图片] [图片] [图片] 1girl, blue_serafuku, blue_skirt, long_sleeves, solo, white_pantyhose, black_gloves, blue_shirt, green_neckerchief, leaf, white_footwear, white_sailor_collar, looking_at_viewer, pleated_skirt, sneakers, bolt_action, knee_pads, sitting, closed_mouth, single_knee_pad, blush, simple_background, white_background, blunt_bangs, bright_pupils, full_body, holding_gun, trash_can
5 15 [图片] [图片] [图片] [图片] [图片] 1girl, blue_shirt, blue_skirt, feet, long_sleeves, no_shoes, soles, solo, toes, white_pantyhose, black_gloves, blue_serafuku, blush, leaf, looking_at_viewer, panties_under_pantyhose, pleated_skirt, green_neckerchief, white_panties, white_sailor_collar, open_mouth, knee_pads, trash_can, simple_background, single_knee_pad, sitting, white_background, ass, blunt_bangs, rifle
6 9 [图片] [图片] [图片] [图片] [图片] 1girl, alternate_costume, playboy_bunny, solo, strapless_leotard, bare_shoulders, detached_collar, looking_at_viewer, blush, rabbit_tail, simple_background, small_breasts, white_background, closed_mouth, white_pantyhose, wrist_cuffs, blue_leotard, bowtie, leaf, ass, fake_tail, no_shoes

表格版本

# 样本数 图片1 图片2 图片3 图片4 图片5 1girl blue_sky cloud day looking_at_viewer navel official_alternate_costume outdoors print_bikini small_breasts stomach straw_hat twin_braids white_bikini blush closed_mouth leaf_print solo ocean seagull sitting white_pupils beach grey_halo hat_on_back twintails simple_background white_background flat_chest nipples leaf completely_nude uncensored open_mouth cleft_of_venus cum_in_pussy cumdrip black_gloves blue_serafuku blue_shirt blue_skirt long_sleeves white_pantyhose white_sailor_collar green_neckerchief pleated_skirt knee_pads white_footwear sneakers bolt_action single_knee_pad blunt_bangs bright_pupils full_body holding_gun trash_can feet no_shoes soles toes panties_under_pantyhose white_panties ass rifle alternate_costume playboy_bunny strapless_leotard bare_shoulders detached_collar rabbit_tail wrist_cuffs blue_leotard bowtie fake_tail
0 5 [图片] [图片] [图片] [图片] [图片] 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 22 [图片] [图片] [图片] [图片] [图片] X X X X X X X X X X X X X X X X X X X
2 7 [图片] [图片] [图片] [图片] [图片] X X X X X X X X X X X X X X X X X
3 10 [图片] [图片] [图片] [图片] [图片] X X X X X X X X X X X X X X X X X X X
4 24 [图片] [图片] [图片] [图片] [图片] 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 15 [图片] [图片] [图片] [图片] [图片] 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 9 [图片] [图片] [图片] [图片] [图片] X X X X X X X X X X X X X X X X X X X X X X
搜集汇总
数据集介绍
main_image_url
构建方式
在二次元角色数据集构建的领域中,CyberHarem/miyu_bluearchive 数据集聚焦于游戏《蔚蓝档案》中的角色“美游”。该数据集通过自动化爬取系统从Danbooru、Pixiv、Zerochan等多个图像托管站点采集原始图像,最终收录了500张高质量图片及其对应的标签信息。构建过程中,系统对图像进行了多尺度处理,提供了原始分辨率版本、短边不超过1200像素的标准化版本,以及经过三阶段裁剪、确保图像区域不小于480×480像素的增强版本,以满足不同训练场景的需求。
特点
该数据集的一个显著特点在于其丰富的标签体系与精细的数据组织方式。核心标签如黑发、长发、红瞳、光环、头顶叶片、呆毛、兽耳、兔耳等被系统性地精简,便于模型聚焦于角色本质特征。此外,数据集提供了基于标签聚类的分析结果,将图像按视觉语义划分为多个簇,每个簇内图像共享相似的属性组合,例如不同服装变体或场景配置,这为研究角色多模态表征提供了结构化视角。
使用方法
使用该数据集时,研究者可通过Hugging Face Hub直接下载不同规格的压缩包。对于需要原始元数据的场景,推荐采用waifuc库加载原始数据集,通过简单的Python代码即可实现图像与标签的迭代访问。具体流程包括从仓库下载dataset-raw.zip文件,解压至本地目录后,利用LocalSource接口逐项读取图像及其文件名和标签信息,从而无缝集成到文本到图像模型的训练或评估流程中。
背景与挑战
背景概述
在文本到图像生成领域,高质量、细粒度的角色数据集是驱动生成模型精准刻画特定人物形象的关键资源。CyberHarem/miyu_bluearchive 数据集由 DeepGHS 团队于近期构建,专注于收录游戏《蔚蓝档案》中的角色“美游”。该数据集通过自动化爬取系统从 Danbooru、Pixiv、Zerochan 等多个主流二次元图像平台采集了 500 张图像,并附带了详尽的标签注释。核心研究问题在于如何为小众动画角色构建一个兼具多样性与标注一致性的训练数据集,以支持后续的扩散模型或生成对抗网络进行角色定制化生成。该数据集的出现为二次元角色生成领域提供了标准化的数据范本,推动了相关社区在角色外貌、服饰及姿态上的可控生成研究。
当前挑战
当前数据集面临的首要挑战在于其规模限制,仅 500 张图像对于训练泛化能力强的生成模型而言仍显不足,可能导致模型在生成动作、视角或复杂背景时出现模式坍塌或细节失真。构建过程中,跨平台数据爬取面临着版权合规与图像质量参差不齐的难题,需通过边缘对齐等预处理手段统一图像尺寸,但部分图像分辨率损失后细节清晰度下降。此外,标签体系虽包含角色核心特征(如兔耳、光环),但不同来源的标签语义存在冲突或缺失,需人工校验与聚类整理。数据集中包含的敏感内容(如裸露图像)亦增加了内容过滤与伦理审查的复杂度,如何在保留数据多样性的同时确保合规性,成为后续扩展与社区共享的重要障碍。
常用场景
经典使用场景
在生成式人工智能与数字艺术交叉领域,该数据集作为特定动漫角色(霞沢美游)的高质量图文配对资源,常被用于文本到图像生成模型的微调与风格对齐任务。研究者通过其包含的500张精心标注图像与标签(如发型、饰品、服装等核心属性),能够有效训练模型理解角色在多种姿态、服饰及场景下的视觉一致性,从而提升生成图像的语义保真度与角色辨识度。
衍生相关工作
该数据集衍生出多项经典工作,包括基于标签聚类分析的服饰迁移研究,以及利用Waifuc框架构建的自动化角色数据集构建管线。此外,其提供的多版本数据(原始、裁剪、标签化)启发了一系列关于数据预处理策略对生成质量影响的对比实验,例如图像长边对齐与区域裁剪策略在细节保持上的权衡分析,为后续角色数据集设计提供了可复现的范例。
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转换的前沿领域,CyberHarem/miyu_bluearchive数据集聚焦于《蔚蓝档案》中角色“美游”的高质量图像与标签集合,为Stable Diffusion等扩散模型的微调与风格化生成提供了精细化的数据支撑。该数据集包含500张经过精心爬取与标注的图像,覆盖了从原始高分辨率到多阶段裁剪的多种规格,并采用标签聚类技术挖掘角色在不同服饰与场景下的视觉特征,如官方泳装、校园制服及兔女郎装扮等。这一工作紧密关联当前AI绘画社区对角色一致性保持与多模态语义对齐的热点需求,通过提供结构化的标签体系与可复用的数据加载工具(如Waifuc),显著降低了二次元角色定制化生成的门槛,对于推动虚拟角色IP的数字化创作与个性化表达具有重要的实践意义与学术价值。
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