GRID 行人重识别数据集

GRID 全称 QMUL underGround Re-IDentification Dataset，包含 250 个行人 image pair，每个 pair 包含同一人的两张图像（由不同视角的摄像机拍摄）。所有图像均拍摄于一个人流量大的地铁站，由 8 个不相临的摄像机捕捉。

GRID, whose full name is QMUL underGround Re-IDentification Dataset, contains 250 pedestrian image pairs. Each pair consists of two images of the same pedestrian captured by cameras with different viewpoints. All images were taken in a busy underground subway station and captured by 8 non-adjacent cameras.

搜集汇总

背景与挑战

背景概述

GRID 行人重识别数据集包含250个行人图像对，每个对由同一人在不同视角摄像机拍摄的两张图像组成，拍摄于地铁站。这些图像在姿势、颜色和光线方面变化大，且分辨率低，画质较差，具有挑战性。