Learning Pose-invariant 3D Object Reconstruction from Single-view Images

by   Bo Peng, et al.

Learning to reconstruct 3D shapes using 2D images is an active research topic, with benefits of not requiring expensive 3D data. However, most work in this direction requires multi-view images for each object instance as training supervision, which oftentimes does not apply in practice. In this paper, we relax the common multi-view assumption and explore a more challenging yet more realistic setup of learning 3D shape from only single-view images. The major difficulty lies in insufficient constraints that can be provided by single view images, which leads to the problem of pose entanglement in learned shape space. As a result, reconstructed shapes vary along input pose and have poor accuracy. We address this problem by taking a novel domain adaptation perspective, and propose an effective adversarial domain confusion method to learn pose-disentangled compact shape space. Experiments on single-view reconstruction show effectiveness in solving pose entanglement, and the proposed method achieves state-of-the-art reconstruction accuracy with high efficiency.


Shape-Aware Human Pose and Shape Reconstruction Using Multi-View Images

We propose a scalable neural network framework to reconstruct the 3D mes...

The importance of silhouette optimization in 3D shape reconstruction system from multiple object scenes

This paper presents a multi stage 3D shape reconstruction system of mult...

Multi-view Consistency as Supervisory Signal for Learning Shape and Pose Prediction

We present a framework for learning single-view shape and pose predictio...

Learning to Reconstruct People in Clothing from a Single RGB Camera

We present a learning-based model to infer the personalized 3D shape of ...

Supervised Learning of the Next-Best-View for 3D Object Reconstruction

Motivated by the advances in 3D sensing technology and the spreading of ...

Learning Single-View 3D Reconstruction with Adversarial Training

Single-view 3D shape reconstruction is an important but challenging prob...

Disjoint Pose and Shape for 3D Face Reconstruction

Existing methods for 3D face reconstruction from a few casually captured...

Code Repositories


The code repository for "Learning Pose-invariant 3D Object Reconstruction from Single-view Images"

view repo

Please sign up or login with your details

Forgot password? Click here to reset