HabitatDyn

HabitatDyn数据集由柏林工业大学创建，包含30个场景，涵盖6种不同速度的动态对象。该数据集通过移动机器人的视角捕捉，包含合成RGB视频、语义标签和深度信息，旨在支持动态对象检测和机器人导航研究。数据集创建过程中，特别关注了机器学习算法与移动机器人应用的结合，以提高机器人在动态环境中的识别和导航能力。

Developed by Technische Universität Berlin, the HabitatDyn dataset consists of 30 scenarios covering dynamic objects with 6 distinct speed levels. Captured from the perspective of a mobile robot, this dataset includes synthetic RGB videos, semantic labels and depth information, and is intended to support research on dynamic object detection and robotic navigation. During its development, particular emphasis was placed on the integration of machine learning algorithms and mobile robotic applications, with the goal of improving robots' recognition and navigation capabilities in dynamic environments.