Skip to main content

OASIS: A Large-Scale Dataset for Single Image 3D in the Wild

Author(s): Chen, Weifang; Qian, Shengyi; Fan, David; Kojima, Noriyuki; Hamilton, Max; et al

Download
To refer to this page use: http://arks.princeton.edu/ark:/88435/pr1nz4j
Abstract: Single-view 3D is the task of recovering 3D properties such as depth and surface normals from a single image. We hypothesize that a major obstacle to single-image 3D is data. We address this issue by presenting Open Annotations of Single Image Surfaces (OASIS), a dataset for single-image 3D in the wild consisting of annotations of detailed 3D geometry for 140,000 images. We train and evaluate leading models on a variety of single-image 3D tasks. We expect OASIS to be a useful resource for 3D vision research. Project site: https://pvl.cs.princeton.edu/OASIS.
Publication Date: 2020
Citation: Chen, Weifeng, Shengyi Qian, David Fan, Noriyuki Kojima, Max Hamilton, and Jia Deng. "OASIS: A Large-Scale Dataset for Single Image 3D in the Wild." IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2020): pp. 676-685. doi:10.1109/CVPR42600.2020.00076
DOI: 10.1109/CVPR42600.2020.00076
ISSN: 1063-6919
EISSN: 2575-7075
Pages: 676 - 685
Type of Material: Conference Article
Journal/Proceeding Title: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Version: Author's manuscript



Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.