PAUL: Procrustean Autoencoder for Unsupervised Lifting
This work addresses 3D reconstruction from 2D images for computer vision applications, representing an incremental improvement over existing methods like Deep NRSfM and C3PDO.
The paper tackles the problem of Non-rigid Structure from Motion (NRSfM) by proposing a 3D deep auto-encoder framework called PAUL, which learns from 2D projections and resolves rigid poses, achieving state-of-the-art performance on benchmarks.
Recent success in casting Non-rigid Structure from Motion (NRSfM) as an unsupervised deep learning problem has raised fundamental questions about what novelty in NRSfM prior could the deep learning offer. In this paper we advocate for a 3D deep auto-encoder framework to be used explicitly as the NRSfM prior. The framework is unique as: (i) it learns the 3D auto-encoder weights solely from 2D projected measurements, and (ii) it is Procrustean in that it jointly resolves the unknown rigid pose for each shape instance. We refer to this architecture as a Procustean Autoencoder for Unsupervised Lifting (PAUL), and demonstrate state-of-the-art performance across a number of benchmarks in comparison to recent innovations such as Deep NRSfM and C3PDO.