Dataset Groups Activity Stream Room-to-Room The Room-to-Room dataset is a photo-realistic dataset for vision-and-language navigation, where agents navigate through indoor environments based on natural language instructions. BibTex: @dataset{Jason_Baldridge_and_Gabriel_Ilharco_and_Vihan_Jain_and_Alexander_Ku_and_Eugene_Ie_2024, abstract = {The Room-to-Room dataset is a photo-realistic dataset for vision-and-language navigation, where agents navigate through indoor environments based on natural language instructions.}, author = {Jason Baldridge and Gabriel Ilharco and Vihan Jain and Alexander Ku and Eugene Ie}, doi = {10.57702/33gdihef}, institution = {No Organization}, keyword = {'Dynamic Time Warping', 'Indoor Environments', 'Instruction Conditioned', 'Natural Language Instructions', 'Natural Language Navigation', 'Navigation', 'Room-to-Room', 'Vision-and-Language Navigation'}, month = {dec}, publisher = {TIB}, title = {Room-to-Room}, url = {https://service.tib.eu/ldmservice/dataset/room-to-room}, year = {2024} }