室内长走廊视觉定位与建图数据

面向基于视觉的机器人环境建模与定位导航基础研究，为突破复杂环境下机器人视觉实时高精度定位技术瓶颈，基于移动视觉机器人平台的双目RGB彩色相机观测，主要记录同济大学校园内的教学楼室内长走廊场景下（变化光照、少弱纹理等复杂条件）的视觉影像，影像数目为200，数据量为131 M。This dataset is intended for fundamental research on vision-based robotic environmental modeling, localization and navigation. To break through the technical bottleneck of real-time high-precision visual localization for robots in complex environments, it collects visual data via binocular RGB color cameras mounted on a mobile vision robotic platform. The dataset mainly records visual images captured in indoor long corridor scenarios of teaching buildings on the campus of Tongji University, under complex conditions such as varying illumination and low texture. The total number of images is 200, with a total data volume of 131 MB.

搜集汇总

背景与挑战

背景概述

该数据集包含同济大学教学楼室内长走廊场景下的200张视觉影像，数据量为131MB，格式为jpg，主要用于机器人视觉定位与建图研究，特别针对复杂光照和少弱纹理等条件。