detection-datasets/fashionpedia_4_categories
收藏资源简介:
Fashionpedia_4_categories数据集是Fashionpedia数据集的一个变体,主要区别在于它只包含4个类别(服装、鞋子、包和配饰),并且创建了新的数据集划分(训练集90%,验证集5%,测试集5%)。该数据集的目的是通过减少类别数量(从46个减少到4个)来简化目标检测任务。数据集使用了`detection_datasets`库进行创建,并通过映射将原始类别合并为4个主要类别,没有对应类别的注释被删除。
The Fashionpedia_4_categories dataset is a variant of the Fashionpedia dataset. Its core differences are twofold: it only includes four categories (clothing, shoes, bags, and accessories), and it features a newly created dataset split with 90% allocated to the training set, 5% to the validation set, and 5% to the test set. The goal of this dataset is to simplify the object detection task by reducing the number of categories from 46 to 4. It was developed using the `detection_datasets` library, where original categories are merged into four main categories via a mapping process, and annotations that do not correspond to any of these categories are removed.
数据集概述:Fashionpedia_4_categories
基本信息
- 名称:Fashionpedia_4_categories
- 任务类别:object-detection
- 语言:英语(en)
- 许可证:CC-BY-4.0
- 多语言性:单语种
- 大小类别:10K<n<100K
- 来源数据集:fashionpedia
- 标签:object-detection, fashion, computer-vision
- paperswithcode_id:fashionpedia
数据集内容
- 类别数量:4
- Clothing
- Shoes
- Bags
- Accessories
- 数据划分:
- 训练集:90%的图像
- 验证集:5%
- 测试集:5%
数据处理
-
映射规则: Python mapping = { shirt, blouse: clothing, top, t-shirt, sweatshirt: clothing, sweater: clothing, cardigan: clothing, jacket: clothing, vest: clothing, pants: clothing, shorts: clothing, skirt: clothing, coat: clothing, dress: clothing, jumpsuit: clothing, cape: clothing, glasses: accessories, hat: accessories, headband, head covering, hair accessory: accessories, tie: accessories, glove: accessories, belt: accessories, tights, stockings: accessories, sock: accessories, shoe: shoes, bag, wallet: bags, scarf: accessories, }
-
结果:无映射对应的类别被移除。




