Toward General Object-level Mapping from Sparse Views with 3D Diffusion Priors

TL;DR

This paper introduces GOM, a system that uses 3D diffusion priors to enable accurate multi-category object mapping from sparse views, overcoming occlusion and noise challenges in robotics applications.

Contribution

GOM leverages a 3D diffusion model as a shape prior with multi-category support and fuses sensor data with diffusion constraints for joint pose and shape estimation.

Findings

GOM outperforms state-of-the-art methods on real-world benchmarks.

It achieves more accurate object mapping from sparse views.

Supports multi-category object mapping without finetuning.

Abstract

Object-level mapping builds a 3D map of objects in a scene with detailed shapes and poses from multi-view sensor observations. Conventional methods struggle to build complete shapes and estimate accurate poses due to partial occlusions and sensor noise. They require dense observations to cover all objects, which is challenging to achieve in robotics trajectories. Recent work introduces generative shape priors for object-level mapping from sparse views, but is limited to single-category objects. In this work, we propose a General Object-level Mapping system, GOM, which leverages a 3D diffusion model as shape prior with multi-category support and outputs Neural Radiance Fields (NeRFs) for both texture and geometry for all objects in a scene. GOM includes an effective formulation to guide a pre-trained diffusion model with extra nonlinear constraints from sensor measurements without finetuning. We also develop a probabilistic optimization formulation to fuse multi-view sensor observations and diffusion priors for joint 3D object pose and shape estimation. Our GOM system demonstrates superior multi-category mapping performance from sparse views, and achieves more accurate mapping results compared to state-of-the-art methods on the real-world benchmarks. We will release our code: https://github.com/TRAILab/GeneralObjectMapping.