mattmdjaga/human_parsing_dataset
收藏资源简介:
该数据集包含17,706张图像和对应的掩码对,用于图像分割任务,具体为语义分割。掩码标签包括背景、帽子、头发、太阳镜、上衣、裙子、裤子、连衣裙、腰带、左鞋、右鞋、脸、左腿、右腿、左臂、右臂、包和围巾等18个类别。数据集的结构包括训练集,数据特征为图像和掩码。数据集的来源是Deep Human Parsing的ATR数据集。
This dataset comprises 17,706 image-mask pairs, tailored for image segmentation tasks, specifically semantic segmentation. The mask labels cover 18 categories: background, hat, hair, sunglasses, upper clothing, skirt, trousers, dress, belt, left shoe, right shoe, face, left leg, right leg, left arm, right arm, bag, and scarf. The dataset includes a training set, with data features consisting of images and their corresponding masks. This dataset is derived from the ATR dataset used in Deep Human Parsing.
数据集卡片:人体解析数据(ATR)
数据集描述
数据集概要
该数据集包含17,706对图像和掩码。掩码标签如下:
- "0": "背景"
- "1": "帽子"
- "2": "头发"
- "3": "太阳镜"
- "4": "上衣"
- "5": "裙子"
- "6": "裤子"
- "7": "连衣裙"
- "8": "腰带"
- "9": "左鞋"
- "10": "右鞋"
- "11": "脸"
- "12": "左腿"
- "13": "右腿"
- "14": "左臂"
- "15": "右臂"
- "16": "包"
- "17": "围巾"
数据集结构
数据实例
[更多信息需要]
数据字段
[更多信息需要]
数据分割
[更多信息需要]
数据集创建
策划理由
[更多信息需要]
源数据
初始数据收集和规范化
[更多信息需要]
源语言生产者是谁?
[更多信息需要]
注释
注释过程
[更多信息需要]
注释者是谁?
[更多信息需要]
个人和敏感信息
[更多信息需要]
使用数据集的注意事项
数据集的社会影响
[更多信息需要]
偏见的讨论
[更多信息需要]
其他已知限制
[更多信息需要]
附加信息
数据集策展人
[更多信息需要]
许可信息
[更多信息需要]
引用信息
bibtex @ARTICLE{ATR, author={Xiaodan Liang and Si Liu and Xiaohui Shen and Jianchao Yang and Luoqi Liu and Jian Dong and Liang Lin and Shuicheng Yan}, journal={Pattern Analysis and Machine Intelligence, IEEE Transactions on}, title={Deep Human Parsing with Active Template Regression}, year={2015}, volume={37}, number={12}, pages={2402-2414}, doi={10.1109/TPAMI.2015.2408360}, ISSN={0162-8828}, month={Dec}}
@InProceedings{CO-CNN, author={Xiaodan Liang and Chunyan Xu and Xiaohui Shen and Jianchao Yang and Si Liu and Jinhui Tang and Liang Lin and Shuicheng Yan}, journal ={Pattern Analysis and Machine Intelligence, IEEE Transactions on}, title={ICCV}, year={2015}, }




