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.

9 pages, 6 figures

References in corpus (4)

• Perspective Transformer Nets: Learning Single-View 3D Object Reconstruction without 3D Supervision
• Hierarchical Surface Prediction for 3D Object Reconstruction
• Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs
• 3D Shape Estimation from 2D Landmarks: A Convex Relaxation Approach

Cited by in corpus (4)

• Image-based 3D Object Reconstruction: State-of-the-Art and Trends in the Deep Learning Era
• PointGrow: Autoregressively Learned Point Cloud Generation with Self-Attention
• Mesh deformation-based single-view 3D reconstruction of thin eyeglasses frames with differentiable rendering
• What Do Single-view 3D Reconstruction Networks Learn?