five

CyberHarem/zara_azurlane

收藏
Hugging Face2024-01-12 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/CyberHarem/zara_azurlane
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 zara/ザラ/扎拉 (Azur Lane) This is the dataset of zara/ザラ/扎拉 (Azur Lane), containing 500 images and their tags. The core tags of this character are `breasts, red_hair, long_hair, bangs, large_breasts, side_ponytail, goggles_on_head, yellow_eyes, ahoge, hair_ornament`, 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 | 901.58 MiB | [Download](https://huggingface.co/datasets/CyberHarem/zara_azurlane/resolve/main/dataset-raw.zip) | Waifuc-Raw | Raw data with meta information (min edge aligned to 1400 if larger). | | 800 | 500 | 439.05 MiB | [Download](https://huggingface.co/datasets/CyberHarem/zara_azurlane/resolve/main/dataset-800.zip) | IMG+TXT | dataset with the shorter side not exceeding 800 pixels. | | stage3-p480-800 | 1326 | 1.01 GiB | [Download](https://huggingface.co/datasets/CyberHarem/zara_azurlane/resolve/main/dataset-stage3-p480-800.zip) | IMG+TXT | 3-stage cropped dataset with the area not less than 480x480 pixels. | | 1200 | 500 | 766.76 MiB | [Download](https://huggingface.co/datasets/CyberHarem/zara_azurlane/resolve/main/dataset-1200.zip) | IMG+TXT | dataset with the shorter side not exceeding 1200 pixels. | | stage3-p480-1200 | 1326 | 1.57 GiB | [Download](https://huggingface.co/datasets/CyberHarem/zara_azurlane/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/zara_azurlane', 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 | 53 | ![](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) | twintails, beret, 1girl, black_headwear, cleavage, dress, looking_at_viewer, blush, white_gloves, solo, cape, smile, thighhighs, italian_flag, epaulettes, red_eyes, skindentation, simple_background | | 1 | 6 | ![](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) | 1boy, 1girl, black_bikini, blush, collarbone, cowgirl_position, girl_on_top, goggles, hetero, looking_at_viewer, o-ring_bikini, sex, solo_focus, wet, bikini_bottom_aside, navel, nipples, vaginal, cleavage, open_mouth, outdoors, penis, pussy, sky, smile, day, mosaic_censoring, pov | | 2 | 8 | ![](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, black_bikini, blush, cleavage, collarbone, looking_at_viewer, o-ring_bikini, smile, solo, goggles, wet, hair_between_eyes, official_alternate_costume, sidelocks, closed_mouth, huge_breasts, skindentation, bare_shoulders, water | | 3 | 8 | ![](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_bikini, blush, cleavage, collarbone, looking_at_viewer, o-ring_bikini, solo, water, wet, official_alternate_costume, poolside, smile, bare_shoulders, navel, pool_ladder, goggles, open_mouth, skindentation, thighs, sidelocks, huge_breasts | | 4 | 35 | ![](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, black_bikini, cleavage, looking_at_viewer, o-ring_bikini, collarbone, goggles, solo, blush, whistle_around_neck, smile, skindentation, wet, navel, multi-strapped_bikini, poolside, mouth_hold, water, official_alternate_costume, pool_ladder, sitting | | 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) | 1girl, black_bikini, blush, cleavage, collarbone, day, goggles, looking_at_viewer, o-ring_bikini, outdoors, skindentation, smile, solo, wet, blue_sky, multi-strapped_bikini, navel, ocean, whistle_around_neck, italian_flag, official_alternate_costume, open_mouth, water, armpits, beach, covered_nipples, huge_breasts, kneeling | | 6 | 13 | ![](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_bikini, cleavage, looking_at_viewer, multi-strapped_bikini, navel, o-ring_bikini, solo, simple_background, white_background, blush, skindentation, smile, wet, collarbone, goggles, official_alternate_costume, sidelocks, bare_shoulders | | 7 | 8 | ![](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) | 1boy, 1girl, black_bikini, blush, collarbone, goggles, hetero, o-ring_bikini, solo_focus, wet, cleavage, paizuri, smile, cum_on_breasts, hair_between_eyes, huge_breasts, looking_at_viewer, pov, breasts_squeezed_together, open_mouth, penis, tongue_out, closed_mouth, nipples, sidelocks, water | ### Table Version | # | Samples | Img-1 | Img-2 | Img-3 | Img-4 | Img-5 | twintails | beret | 1girl | black_headwear | cleavage | dress | looking_at_viewer | blush | white_gloves | solo | cape | smile | thighhighs | italian_flag | epaulettes | red_eyes | skindentation | simple_background | 1boy | black_bikini | collarbone | cowgirl_position | girl_on_top | goggles | hetero | o-ring_bikini | sex | solo_focus | wet | bikini_bottom_aside | navel | nipples | vaginal | open_mouth | outdoors | penis | pussy | sky | day | mosaic_censoring | pov | hair_between_eyes | official_alternate_costume | sidelocks | closed_mouth | huge_breasts | bare_shoulders | water | poolside | pool_ladder | thighs | whistle_around_neck | multi-strapped_bikini | mouth_hold | sitting | blue_sky | ocean | armpits | beach | covered_nipples | kneeling | white_background | paizuri | cum_on_breasts | breasts_squeezed_together | tongue_out | |----:|----------:|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:--------------------------------|:------------|:--------|:--------|:-----------------|:-----------|:--------|:--------------------|:--------|:---------------|:-------|:-------|:--------|:-------------|:---------------|:-------------|:-----------|:----------------|:--------------------|:-------|:---------------|:-------------|:-------------------|:--------------|:----------|:---------|:----------------|:------|:-------------|:------|:----------------------|:--------|:----------|:----------|:-------------|:-----------|:--------|:--------|:------|:------|:-------------------|:------|:--------------------|:-----------------------------|:------------|:---------------|:---------------|:-----------------|:--------|:-----------|:--------------|:---------|:----------------------|:------------------------|:-------------|:----------|:-----------|:--------|:----------|:--------|:------------------|:-----------|:-------------------|:----------|:-----------------|:----------------------------|:-------------| | 0 | 53 | ![](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 | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1 | 6 | ![](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 | X | X | X | X | X | X | X | | | | | | | | | | | | | | | | | | | | | | | | | | | 2 | 8 | ![](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 | | | | | | | | | | | | | | | | | | | | 3 | 8 | ![](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 | | | | | | | | | | | | | | | | | 4 | 35 | ![](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 | | | | | | | | | | | | | 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 | | | | | | | 6 | 13 | ![](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 | | | | | | 7 | 8 | ![](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 | 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

  • 名称: zara/ザラ/扎拉 (Azur Lane)

数据集内容

  • 内容: 包含500张图像及其标签。
  • 核心标签: 包括breasts, red_hair, long_hair, bangs, large_breasts, side_ponytail, goggles_on_head, yellow_eyes, ahoge, hair_ornament

数据集来源

  • 来源: 图像从多个网站爬取,如danbooru, pixiv, zerochan等。
  • 爬取系统: 由DeepGHS Team提供技术支持。

数据集版本与下载

版本列表

名称 图像数量 大小 下载链接 类型 描述
raw 500 901.58 MiB 下载 Waifuc-Raw 包含元信息的原始数据(最小边对齐到1400像素)。
800 500 439.05 MiB 下载 IMG+TXT 短边不超过800像素的图像数据集。
stage3-p480-800 1326 1.01 GiB 下载 IMG+TXT 三阶段裁剪,区域不小于480x480像素的数据集。
1200 500 766.76 MiB 下载 IMG+TXT 短边不超过1200像素的图像数据集。
stage3-p480-1200 1326 1.57 GiB 下载 IMG+TXT 三阶段裁剪,区域不小于480x480像素的数据集。

数据集加载

  • 加载工具: 使用waifuc
  • 加载代码示例: 提供Python代码用于下载和加载原始数据集。

数据集标签集群

集群列表

# 样本数 图像示例 标签
0 53 twintails, beret, 1girl, black_headwear, cleavage, dress, looking_at_viewer, blush, white_gloves, solo, cape, smile, thighhighs, italian_flag, epaulettes, red_eyes, skindentation, simple_background
1 6 1boy, 1girl, black_bikini, blush, collarbone, cowgirl_position, girl_on_top, goggles, hetero, looking_at_viewer, o-ring_bikini, sex, solo_focus, wet, bikini_bottom_aside, navel, nipples, vaginal, cleavage, open_mouth, outdoors, penis, pussy, sky, day, mosaic_censoring, pov
2 8 1girl, black_bikini, blush, cleavage, collarbone, looking_at_viewer, o-ring_bikini, smile, solo, goggles, wet, hair_between_eyes, official_alternate_costume, sidelocks, closed_mouth, huge_breasts, skindentation, bare_shoulders, water
3 8 1girl, black_bikini, blush, cleavage, collarbone, looking_at_viewer, o-ring_bikini, solo, water, wet, official_alternate_costume, poolside, smile, bare_shoulders, navel, pool_ladder, goggles, open_mouth, skindentation, thighs, sidelocks, huge_breasts
4 35 1girl, black_bikini, cleavage, looking_at_viewer, o-ring_bikini, collarbone, goggles, solo, blush, whistle_around_neck, smile, skindentation, wet, navel, multi-strapped_bikini, poolside, mouth_hold, water, official_alternate_costume, pool_ladder, sitting
5 5 1girl, black_bikini, blush, cleavage, collarbone, day, goggles, looking_at_viewer, o-ring_bikini, outdoors, skindentation, smile, solo, wet, blue_sky, multi-strapped_bikini, navel, ocean, whistle_around_neck, italian_flag, official_alternate_costume, open_mouth, water, armpits, beach, covered_nipples, huge_breasts, kneeling
6 13 1girl, black_bikini, cleavage, looking_at_viewer, multi-strapped_bikini, navel, o-ring_bikini, solo, simple_background, white_background, blush, skindentation, smile, wet, collarbone, goggles, official_alternate_costume, sidelocks, bare_shoulders
7 8 1boy, 1girl, black_bikini, blush, collarbone, goggles, hetero, o-ring_bikini, solo_focus, wet, cleavage, paizuri, smile, cum_on_breasts, hair_between_eyes, huge_breasts, looking_at_viewer, pov, breasts_squeezed_together, open_mouth, penis, tongue_out, closed_mouth, nipples, sidelocks, water

数据集许可

  • 许可: MIT
搜集汇总
数据集介绍
main_image_url
构建方式
在二次元图像生成领域,角色数据集的质量与多样性直接影响模型的表现。该数据集聚焦于《碧蓝航线》中的角色扎拉(Zara),通过自动化爬取系统从Danbooru、Pixiv、Zerochan等多个知名图像平台采集原始图像,并由DeepGHS团队提供技术支撑。采集后的图像经过严格筛选,最终保留了500张高质量图片,并依据角色核心标签(如红发、长发、大胸、侧马尾等)进行了标签精简。数据集提供了多种预处理版本,包括原始元数据包、短边限制为800或1200像素的标准化版本,以及经过三阶段裁剪、确保区域不小于480×480像素的增强版本,以适应不同的训练需求。
特点
该数据集的核心特点在于其精细化的标签体系与多版本设计。每张图像均附有详细的描述标签,不仅涵盖了角色的基础特征,还通过聚类分析将图像按服装、姿势和场景分组,例如泳装主题的簇群揭示了角色在不同情境下的表现模式。数据集的版本多样性是其另一大亮点,从保留完整元数据的原始包到压缩后的标准化图像,再到多阶段裁剪的增强数据,为不同规模的模型训练提供了灵活的选择。此外,数据集还提供了标签聚类结果的文本和表格形式,便于研究者挖掘角色在不同装扮下的视觉规律。
使用方法
研究者可通过Hugging Face Hub直接下载数据集,并利用waifuc库高效加载原始数据。具体操作为:首先使用huggingface_hub库下载dataset-raw.zip压缩包,解压至指定目录后,通过LocalSource接口读取图像及其对应的文件名和标签信息。对于需要特定尺寸或裁剪版本的场景,可直接下载对应的打包文件(如800或stage3-p480-800版本),这些包采用IMG+TXT格式,即图像与标签文件配对存储,便于直接集成到常见的文本到图像训练流程中。数据集的聚类结果以图像和表格形式呈现,可用于分析角色在不同情境下的标签分布,辅助数据增强或风格迁移研究。
背景与挑战
背景概述
在人工智能生成内容(AIGC)领域,文本到图像(text-to-image)模型的发展催生了大量针对特定角色或风格的专业化数据集。CyberHarem/zara_azurlane数据集由DeepGHS团队于近期创建,聚焦于《碧蓝航线》中的角色“扎拉”(Zara)。该数据集包含500张精心采集的图像及其对应标签,核心特征涵盖红发、长发、侧马尾等视觉元素。研究团队通过自动化爬虫系统从Danbooru、Pixiv、Zerochan等多个平台收集数据,旨在为二次元角色生成任务提供高质量的训练素材。该数据集不仅服务于角色定制化生成,还通过多分辨率版本(如800px、1200px)和三阶段裁剪方案,为不同计算资源的研究者提供灵活选择,对动漫风格图像生成领域具有重要参考价值。
当前挑战
该数据集面临的核心挑战包括:1)领域问题层面,二次元角色生成需解决细粒度视觉特征保真度问题,例如扎拉标志性的侧马尾、护目镜等细节在生成结果中易出现形变或丢失,当前500张样本量难以覆盖所有姿态与光照条件下的特征分布。2)构建过程中,数据来源的多样性导致标签一致性难以保证,不同平台(如Danbooru与Pixiv)的标注体系存在差异,需通过核心标签剪枝策略进行标准化处理。此外,自动爬取系统可能引入低质量或重复图像,需人工校验以剔除噪声。多分辨率版本(如stage3-p480-800)的裁剪方案虽提升了适用性,但过度裁剪可能导致关键构图元素(如全身服装细节)丢失,需在数据增强与完整性间寻求平衡。
常用场景
经典使用场景
CyberHarem/zara_azurlane数据集以《碧蓝航线》中角色扎拉为核心,汇集了500张经过精细标注的二次元图像,涵盖多种姿态、服装与场景变体。该数据集最经典的使用场景在于驱动基于文本到图像生成模型的微调与个性化训练,尤其适用于扩散模型(如Stable Diffusion)的定制化角色生成任务。通过提供短边为800或1200像素的标准化图像与标签对,研究者可高效地训练模型学习扎拉独特的视觉特征,从而在保持角色一致性的前提下,实现从文本描述到高质量图像的精准映射,为二次元角色生成领域提供了标准化的数据基石。
衍生相关工作
围绕该数据集,衍生出多项推动领域发展的经典工作。其中,基于Waifuc框架构建的自动化数据采集与标注流水线,为其他动漫角色数据集(如碧蓝航线系列角色)的规模化构建提供了技术范式。研究者利用该数据集开发的角色一致性保持算法,在文本到图像生成中显著提升了角色面部特征、服饰细节的稳定性。此外,数据集中标签聚类分析揭示的装备与场景共现模式,启发了多模态学习中的细粒度属性解耦研究,进一步催生了面向二次元图像的风格迁移与姿态生成等前沿探索。
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定制领域,基于扩散模型的文本到图像技术正迎来爆发式增长。该数据集聚焦于《碧蓝航线》中的热门角色扎拉,通过整合来自Danbooru、Pixiv等多平台的500张高质量图像与精细标注,为角色一致性生成提供了关键支撑。前沿研究方向包括利用多尺度裁剪(如480-1200像素三阶段处理)提升细节保真度,以及通过标签聚类挖掘角色在不同着装(如官方泳装、日常便服)下的视觉特征,从而推动可控角色换装、姿态迁移和跨模态生成等热点应用。该数据集不仅降低了二次元AI创作的数据门槛,还为解决小样本角色定制中的过拟合与风格泛化问题提供了宝贵基准,对虚拟偶像产业与游戏资产自动化生产具有深远意义。
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