Q-SLAM: Quadric Representations for Monocular SLAM

TL;DR

Q-SLAM introduces quadric surface representations to improve monocular SLAM by enhancing depth accuracy and scene modeling efficiency through quadric-based decomposition and a novel transformer architecture.

Contribution

The paper proposes a new quadric-based volumetric representation and a quadric-decomposed transformer for improved monocular SLAM performance.

Findings

Significantly improves depth estimation accuracy.

Achieves superior performance over depth-estimation reliant methods.

Comparable accuracy to ground-truth depth methods.

Abstract

In this paper, we reimagine volumetric representations through the lens of quadrics. We posit that rigid scene components can be effectively decomposed into quadric surfaces. Leveraging this assumption, we reshape the volumetric representations with million of cubes by several quadric planes, which results in more accurate and efficient modeling of 3D scenes in SLAM contexts. First, we use the quadric assumption to rectify noisy depth estimations from RGB inputs. This step significantly improves depth estimation accuracy, and allows us to efficiently sample ray points around quadric planes instead of the entire volume space in previous NeRF-SLAM systems. Second, we introduce a novel quadric-decomposed transformer to aggregate information across quadrics. The quadric semantics are not only explicitly used for depth correction and scene decomposition, but also serve as an implicit supervision signal for the mapping network. Through rigorous experimental evaluation, our method exhibits superior performance over other approaches relying on estimated depth, and achieves comparable accuracy to methods utilizing ground truth depth on both synthetic and real-world datasets.