AI2-THOR

doi:doi:10.57702/ga4hm35q

AI2-THOR

The dataset used in the paper is the AI2-THOR dataset, which is a 3D indoor environment dataset. The agent must navigate from a random starting position to a pre-specified target location only via egocentric view. The dataset contains 4 different room categories: kitchens, living rooms, bedrooms, and bathrooms.

BibTex:

@dataset{Nuo_Xu_and_Wen_Wang_and_Rong_Yang_and_Mengjie_Qin_and_Zheyuan_Lin_and_Wei_Song_and_Chunlong_Zhang_and_Jason_Gu_and_Chao_Li_2024,
    abstract = {The dataset used in the paper is the AI2-THOR dataset, which is a 3D indoor environment dataset. The agent must navigate from a random starting position to a pre-specified target location only via egocentric view. The dataset contains 4 different room categories: kitchens, living rooms, bedrooms, and bathrooms.},
    author = {Nuo Xu and Wen Wang and Rong Yang and Mengjie Qin and Zheyuan Lin and Wei Song and Chunlong Zhang and Jason Gu and Chao Li},
    doi = {10.57702/ga4hm35q},
    institution = {No Organization},
    keyword = {'3D Indoor Environment', '3D indoor environment', 'Embodied AI', 'Navigation', 'Object Goal Navigation', 'Robotics', 'Task Completion', 'Virtual Environment', 'Visual AI'},
    month = {dec},
    publisher = {TIB},
    title = {AI2-THOR},
    url = {https://service.tib.eu/ldmservice/dataset/ai2-thor},
    year = {2024}
}