In-Shop (In-shop Clothes Retrieval Benchmark)

In-shop Clothes Retrieval Benchmark 评估店内 Clothes Retrieval 的表现。这是 DeepFashion 的一个大子集，包含大的姿势和尺度变化。它还具有大多样性、大数量和丰富的注释，包括：7,982个服装项目； 52,712 张店内服装图片和约 200,000 个交叉姿势/比例对；每个图像都由边界框、服装类型和姿势类型进行注释。

The In-shop Clothes Retrieval Benchmark evaluates the performance of in-shop clothes retrieval. It is a large subset of DeepFashion, featuring significant variations in pose and scale. The benchmark also boasts high diversity, a large volume of data, and rich annotations, including: 7,982 clothing items; 52,712 in-shop clothing images, and approximately 200,000 cross-pose/scale pairs; each image is annotated with bounding boxes, clothing categories, and pose types.

搜集汇总

背景与挑战

背景概述

In-Shop是一个专注于店内服装检索的大规模数据集，包含7,982个服装项目和52,712张图片，每张图片都有详细的注释（如边界框、服装类型和姿势类型）。该数据集具有高度多样性和丰富的注释，适用于服装识别和检索任务。