Interactive Gibson Environment (iGibson 0.5)

Interactive Gibson Environment (iGibson 0.5) 是由斯坦福大学开发的一个模拟环境数据集，专注于交互式导航任务。该数据集包含106个场景，其中1984个对象可以进行交互，涵盖椅子、桌子、门等五类常见室内物品。数据集通过结合自动算法和人工标注，将原始模型替换为可交互的高分辨率CAD模型，以保持视觉真实性。该数据集旨在为机器人学领域的研究人员提供一个可控且可重复的测试平台，用于开发和评估交互式导航解决方案，特别是在复杂和杂乱的室内环境中。

Interactive Gibson Environment (iGibson 0.5) is a simulated environment dataset developed by Stanford University, focusing on interactive navigation tasks. The dataset contains 106 scenes, with 1984 interactable objects covering five common indoor item categories such as chairs, tables, and doors. By combining automated algorithms and manual annotation, the dataset replaces original models with interactable high-resolution CAD models to maintain visual realism. It aims to provide researchers in the field of robotics with a controllable and reproducible testbed for developing and evaluating interactive navigation solutions, particularly in complex and cluttered indoor environments.

搜集汇总

背景与挑战

背景概述

Interactive Gibson Environment (iGibson 0.5) 是由斯坦福大学开发的模拟环境数据集，专注于交互式导航任务，包含106个场景和1984个可交互对象，涵盖五类常见室内物品。该数据集通过高分辨率CAD模型保持视觉真实性，旨在为机器人学研究提供一个可控测试平台，用于评估复杂室内环境中的导航解决方案。