TY - GEN
T1 - Matterport3D
T2 - 7th IEEE International Conference on 3D Vision, 3DV 2017
AU - Chang, Angel
AU - Dai, Angela
AU - Funkhouser, Thomas
AU - Halber, Maciej
AU - Niebner, Matthias
AU - Savva, Manolis
AU - Song, Shuran
AU - Zeng, Andy
AU - Zhang, Yinda
N1 - Funding Information:
Development of tools for processing the data were supported by Google Tango, Intel, Facebook, NSF (IIS-1251217 and VEC 1539014/1539099)
Publisher Copyright:
© 2017 IEEE.
PY - 2018/5/25
Y1 - 2018/5/25
N2 - Access to large, diverse RGB-D datasets is critical for training RGB-D scene understanding algorithms. However, existing datasets still cover only a limited number of views or a restricted scale of spaces. In this paper, we introduce Matterport3D, a large-scale RGB-D dataset containing 10,800 panoramic views from 194,400 RGB-D images of 90 building-scale scenes. Annotations are provided with surface reconstructions, camera poses, and 2D and 3D semantic segmentations. The precise global alignment and comprehensive, diverse panoramic set of views over entire buildings enable a variety of supervised and self-supervised computer vision tasks, including keypoint matching, view overlap prediction, normal prediction from color, semantic segmentation, and region classification.
AB - Access to large, diverse RGB-D datasets is critical for training RGB-D scene understanding algorithms. However, existing datasets still cover only a limited number of views or a restricted scale of spaces. In this paper, we introduce Matterport3D, a large-scale RGB-D dataset containing 10,800 panoramic views from 194,400 RGB-D images of 90 building-scale scenes. Annotations are provided with surface reconstructions, camera poses, and 2D and 3D semantic segmentations. The precise global alignment and comprehensive, diverse panoramic set of views over entire buildings enable a variety of supervised and self-supervised computer vision tasks, including keypoint matching, view overlap prediction, normal prediction from color, semantic segmentation, and region classification.
UR - http://www.scopus.com/inward/record.url?scp=85043374004&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85043374004&partnerID=8YFLogxK
U2 - 10.1109/3DV.2017.00081
DO - 10.1109/3DV.2017.00081
M3 - Conference contribution
AN - SCOPUS:85043374004
T3 - Proceedings - 2017 International Conference on 3D Vision, 3DV 2017
SP - 667
EP - 676
BT - Proceedings - 2017 International Conference on 3D Vision, 3DV 2017
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 10 October 2017 through 12 October 2017
ER -