深度学习为何能提升视觉同步定位与地图构建？

中文摘要 AI 辅助

视觉同步定位与地图构建（Visual SLAM）是一项在众多实际应用中广泛使用的成熟技术。然而，在诸如低纹理、严重运动模糊和光照不佳等具有挑战性的视觉条件下，其性能仍会下降。基于深度学习的系统优于传统基于几何的系统，通过在循环架构中将学习到的二维数据关联和不确定性与可微几何优化相结合，取得了当前最优结果。但究竟哪些组件是这一成功的根本原因仍不明确。本文探讨基于深度学习的系统的卓越性能主要是由学习到的二维数据关联、学习到的二维数据关联与不确定性的组合，还是循环架构本身驱动的。我们通过对照研究进行实证调查。结果表明，基于深度学习的视觉同步定位与地图构建系统的成功取决于学习到的二维数据关联和不确定性，而非其循环架构，这凸显了基于学习的范式在设计这些组件方面的必要性。论文录用后，代码将作为开源发布。

英文摘要

Visual SLAM is a well-established technology utilized in a wide range of real-world applications. However, its performance still degrades under challenging visual conditions, such as low texture, severe motion blur, and poor illumination. Systems based on deep learning outperform classical geometry-based ones and achieve state-of-the-art results by combining learned 2D data association and uncertainty with differentiable geometric optimization in recurrent architectures. Still, it remains unclear exactly which components are fundamentally responsible for this success. In this paper, we ask: Is the superior performance of deep learning-based systems driven primarily by learned 2D data association, the combination of learned 2D data association and uncertainty, or the recurrent architecture itself? We investigate this question empirically by conducting a controlled study. Our findings reveal that the success of DL-based V-SLAM systems hinges on learned 2D data association and uncertainty rather than their recurrent architecture, underscoring the necessity of learning-based paradigms for the design of these components.

发表机构

• Robotics and Perception Group, Department of Informatics, University of Zurich（苏黎世大学信息学系机器人与感知组）

机构由 AI 辅助整理，请以论文原文为准。