A Versatile Scene Model with Differentiable Visibility Applied to Generative Pose Estimation

TL;DR

This paper introduces a novel scene representation with a differentiable visibility model that improves the accuracy and efficiency of generative pose estimation by better handling occlusions through a smooth, closed-form formulation.

Contribution

The authors propose a new scene model that provides an analytically differentiable visibility function, enabling more effective optimization in generative pose estimation tasks.

Findings

Enhanced convergence in pose optimization tasks

Improved occlusion handling in 3D reconstruction

Versatile application across multiple pose estimation problems

Abstract

Generative reconstruction methods compute the 3D configuration (such as pose and/or geometry) of a shape by optimizing the overlap of the projected 3D shape model with images. Proper handling of occlusions is a big challenge, since the visibility function that indicates if a surface point is seen from a camera can often not be formulated in closed form, and is in general discrete and non-differentiable at occlusion boundaries. We present a new scene representation that enables an analytically differentiable closed-form formulation of surface visibility. In contrast to previous methods, this yields smooth, analytically differentiable, and efficient to optimize pose similarity energies with rigorous occlusion handling, fewer local minima, and experimentally verified improved convergence of numerical optimization. The underlying idea is a new image formation model that represents opaque objects by a translucent medium with a smooth Gaussian density distribution which turns visibility into a smooth phenomenon. We demonstrate the advantages of our versatile scene model in several generative pose estimation problems, namely marker-less multi-object pose estimation, marker-less human motion capture with few cameras, and image-based 3D geometry estimation.