Single-view Object Shape Reconstruction Using Deep Shape Prior and Silhouette
This addresses the ill-posed problem of single-view 3D reconstruction for computer vision applications, but it is incremental as it builds on traditional optimization approaches with deep learning enhancements.
The paper tackles 3D shape reconstruction from a single image by using an online optimization framework with a deep shape prior and silhouette cues, achieving comparable performance to state-of-the-art methods without end-to-end network training.
3D shape reconstruction from a single image is a highly ill-posed problem. Modern deep learning based systems try to solve this problem by learning an end-to-end mapping from image to shape via a deep network. In this paper, we aim to solve this problem via an online optimization framework inspired by traditional methods. Our framework employs a deep autoencoder to learn a set of latent codes of 3D object shapes, which are fitted by a probabilistic shape prior using Gaussian Mixture Model (GMM). At inference, the shape and pose are jointly optimized guided by both image cues and deep shape prior without relying on an initialization from any trained deep nets. Surprisingly, our method achieves comparable performance to state-of-the-art methods even without training an end-to-end network, which shows a promising step in this direction.