3DP3: 3D Scene Perception via Probabilistic Programming

TL;DR

3DP3 is a probabilistic programming framework that infers detailed 3D scene structure and object poses from RGB-D images, outperforming deep learning methods in accuracy and generalization.

Contribution

The paper introduces 3DP3, a novel inverse graphics framework combining voxel models, scene graphs, and depth likelihoods for improved 3D scene understanding.

Findings

More accurate 6DoF object pose estimation than deep learning baselines

Better generalization to challenging scenes with novel viewpoints and occlusion

Effective inference of scene structure including object contacts and poses

Abstract

We present 3DP3, a framework for inverse graphics that uses inference in a structured generative model of objects, scenes, and images. 3DP3 uses (i) voxel models to represent the 3D shape of objects, (ii) hierarchical scene graphs to decompose scenes into objects and the contacts between them, and (iii) depth image likelihoods based on real-time graphics. Given an observed RGB-D image, 3DP3's inference algorithm infers the underlying latent 3D scene, including the object poses and a parsimonious joint parametrization of these poses, using fast bottom-up pose proposals, novel involutive MCMC updates of the scene graph structure, and, optionally, neural object detectors and pose estimators. We show that 3DP3 enables scene understanding that is aware of 3D shape, occlusion, and contact structure. Our results demonstrate that 3DP3 is more accurate at 6DoF object pose estimation from real images than deep learning baselines and shows better generalization to challenging scenes with novel viewpoints, contact, and partial observability.