ELLIPSDF: Joint Object Pose and Shape Optimization with a Bi-level Ellipsoid and Signed Distance Function Description

TL;DR

This paper introduces ELLIPSDF, a compact bi-level model for joint object pose and shape optimization from multi-view RGB-D data, enabling detailed and efficient object mapping for autonomous systems.

Contribution

It presents a novel bi-level object model combining coarse scale and fine shape details, with an optimization algorithm for accurate object map inference from RGB-D observations.

Findings

Outperforms state-of-the-art methods on ScanNet dataset

Efficiently captures object identities, poses, and shapes

Enables onboard storage of large multi-category object maps

Abstract

Autonomous systems need to understand the semantics and geometry of their surroundings in order to comprehend and safely execute object-level task specifications. This paper proposes an expressive yet compact model for joint object pose and shape optimization, and an associated optimization algorithm to infer an object-level map from multi-view RGB-D camera observations. The model is expressive because it captures the identities, positions, orientations, and shapes of objects in the environment. It is compact because it relies on a low-dimensional latent representation of implicit object shape, allowing onboard storage of large multi-category object maps. Different from other works that rely on a single object representation format, our approach has a bi-level object model that captures both the coarse level scale as well as the fine level shape details. Our approach is evaluated on the large-scale real-world ScanNet dataset and compared against state-of-the-art methods.