Compact model representation for 3D reconstruction

Abstract

3D reconstruction from 2D images is a central problem in computer vision. Recent works have been focusing on reconstruction directly from a single image. It is well known however that only one image cannot provide enough information for such a reconstruction. A prior knowledge that has been entertained are 3D CAD models due to its online ubiquity. A fundamental question is how to compactly represent millions of CAD models while allowing generalization to new unseen objects with fine-scaled geometry. We introduce an approach to compactly represent a 3D mesh. Our method first selects a 3D model from a graph structure by using a novel free-form deformation (FFD) 3D-2D registration, and then the selected 3D model is refined to best fit the image silhouette. We perform a comprehensive quantitative and qualitative analysis that demonstrates impressive dense and realistic 3D reconstruction from single images.

Publication
In International Conference on 3D Vision (3DV 2017)

Drawing Drawing

Example of a 3D CAD model being fit to a single image using our FFD anchor registration and silhouette fitting.

Drawing Drawing Drawing

Example of our 3D reconstruction from a single image.

Drawing Drawing Drawing

[1] C. Kong, C-H. Lin and S. Lucey, Using Locally Corresponding CAD Models for Dense 3D Reconstructions from a Single Image, CVPR 2017.