We consider the problem of estimating detailed 3-d structure from a single still image of an unstructured environment. Our goal is to create 3-d models which are both quantitatively accurate as well as visually pleasing.
For each small homogeneous patch in the image, we use a Markov Random Field (MRF) to infer a set of "plane parameters" that capture both the 3-d location and 3-d orientation of the patch. The MRF, trained via supervised learning, models both image depth cues as well as the relationships between different parts of the image. Other than assuming that the environment is made up of a number of small planes, our model makes no explicit assumptions about the structure of the scene; this enables the algorithm to capture much more detailed 3-d structure than does prior art, and also give a much richer experience in the 3-d flythroughs created using image-based rendering, even for scenes with significant non-vertical structure.
Using this approach, we have created qualitatively correct 3-d models for 64.9% of 588 images downloaded from the internet. We have also extended our model to produce large scale 3d models from a few images.
ResultsTraining DataDetailed Results on 588+134 images (Nov 2006) Make 3D model from your imagehttp://make3d.stanford.edu(In two simple steps: upload and browse-in-3d !) Publications
Learning 3-D Scene Structure from a Single Still Image,
3-D Reconstruction from Sparse Views using Monocular Vision,
3-D Depth Reconstruction from a Single Still Image,
LinksPeople: Ashutosh Saxena, Min Sun, Andrew Y. NgReconstruction3d group Wiki Monocular Depth Estimation Improving Stereo-vision Autonomous driving using monocular vision Indoor single image 3-d reconstruction (More) Outdoor single image "popups" |