OASIS: A Large-Scale Dataset for Single Image 3D in the Wild
- 1 June 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 676-685
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.Keywords
All Related Versions
This publication has 31 references indexed in Scilit:
- Designing deep networks for surface normal estimation2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Vision meets robotics: The KITTI datasetThe International Journal of Robotics Research, 2013
- OpenSurfacesACM Transactions on Graphics, 2013
- Boundary Cues for 3D Object Shape Recovery2013
- Color Constancy, Intrinsic Images, and Shape Estimation2012
- Recovering Occlusion Boundaries from an ImageInternational Journal of Computer Vision, 2010
- Contour Detection and Hierarchical Image SegmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Occlusion Boundaries from Motion: Low-Level Detection and Mid-Level ReasoningInternational Journal of Computer Vision, 2009
- 3-D Depth Reconstruction from a Single Still ImageInternational Journal of Computer Vision, 2007