Object-Centric Photometric Bundle Adjustment with Deep Shape Prior

TL;DR

This paper integrates deep shape priors into photometric bundle adjustment to improve 3D shape reconstruction from image sequences, combining geometric and learned shape information for more reliable results.

Contribution

It introduces a novel method that incorporates deep shape generators into photometric bundle adjustment, bridging the gap between classical SfM and deep learning approaches.

Findings

Enhanced 3D shape reconstruction accuracy

Effective integration of shape priors into optimization framework

Impressive results demonstrating the method's potential

Abstract

Reconstructing 3D shapes from a sequence of images has long been a problem of interest in computer vision. Classical Structure from Motion (SfM) methods have attempted to solve this problem through projected point displacement \& bundle adjustment. More recently, deep methods have attempted to solve this problem by directly learning a relationship between geometry and appearance. There is, however, a significant gap between these two strategies. SfM tackles the problem from purely a geometric perspective, taking no account of the object shape prior. Modern deep methods more often throw away geometric constraints altogether, rendering the results unreliable. In this paper we make an effort to bring these two seemingly disparate strategies together. We introduce learned shape prior in the form of deep shape generators into Photometric Bundle Adjustment (PBA) and propose to accommodate full 3D shape generated by the shape prior within the optimization-based inference framework, demonstrating impressive results.