[论文解读] InfiniTAM v3: A Framework for Large-Scale 3D Reconstruction with Loop Closure

Victor Adrian Prisacariu, Olaf Kähler+5 more|arXiv (Cornell University)|Aug 2, 2017

Robotics and Sensor-Based Localization参考文献 7被引用 86

一句话总结

InfiniTAM v3 提出了一种可扩展的 3D 重建框架，使用基于哈希的体素哈希、鲁棒跟踪、随机 Ferns 重定位器，以及全局一致的子图优化以实现回环闭合。

ABSTRACT

Volumetric models have become a popular representation for 3D scenes in recent years. One breakthrough leading to their popularity was KinectFusion, which focuses on 3D reconstruction using RGB-D sensors. However, monocular SLAM has since also been tackled with very similar approaches. Representing the reconstruction volumetrically as a TSDF leads to most of the simplicity and efficiency that can be achieved with GPU implementations of these systems. However, this representation is memory-intensive and limits applicability to small-scale reconstructions. Several avenues have been explored to overcome this. With the aim of summarizing them and providing for a fast, flexible 3D reconstruction pipeline, we propose a new, unifying framework called InfiniTAM. The idea is that steps like camera tracking, scene representation and integration of new data can easily be replaced and adapted to the user's needs. This report describes the technical implementation details of InfiniTAM v3, the third version of our InfiniTAM system. We have added various new features, as well as making numerous enhancements to the low-level code that significantly improve our camera tracking performance. The new features that we expect to be of most interest are (i) a robust camera tracking module; (ii) an implementation of Glocker et al.'s keyframe-based random ferns camera relocaliser; (iii) a novel approach to globally-consistent TSDF-based reconstruction, based on dividing the scene into rigid submaps and optimising the relative poses between them; and (iv) an implementation of Keller et al.'s surfel-based reconstruction approach.

研究动机与目标

• 将 KinectFusion 风格的 3D 重建提升到大尺度环境，借助基于哈希的体积表示。
• 提高相机跟踪的鲁棒性，并在跟踪失败时加入重定位。
• 通过子图和子图间约束实现全局一致的重建。
• 在同一框架内提供基于 surfel 的重建作为可选替代方案。

提出的方法

• 使用哈希表将体素块映射到全局体素块数组，从而实现可扩展的 3D 表示。
• 构建跨设备的引擎架构，包含抽象层、设备特定层和设备无关层，以提高模块性和性能。
• 实现每帧六阶段处理：跟踪、分配、换入、集成、射线投射、换出。
• 使用 ITMDepthTracker、ITMColorTracker 和 ITMExtendedTracker 提升跟踪能力，包含 RGB-D 和光度项。
• 引入随机 Ferns 重定位器，用于相机位姿重定位和回环检测。
• 通过子图和子图间约束优化提供全局一致的重建（多引擎管线）。

实验结果

研究问题

• RQ1 InfiniTAM 如何利用体素哈希将 KinectFusion 风格的重建扩展到大场景？
• RQ2 如何提高跟踪的鲁棒性，使其在跟踪失败时能够进行重定位？
• RQ3 如何在大规模重建中检测回环并实现全局一致性？
• RQ4 在同一框架内，基于 TSDF 与基于 surfel 的表示之间有何权衡？
• RQ5 如何组织跨设备架构以在 CPU、GPU 和其他加速器上优化性能？

主要发现

• 已开发并集成了鲁棒跟踪模块与重定位器，以在跟踪失败后恢复位姿。
• 引入基于子图及子图间约束的全局一致重建方法。
• 在 InfiniTAM v3 中实现基于 surfel 的重建，作为 TSDF 的可选替代方案。
• 基于哈希表的体素哈希可扩展到大场景，同时通过换出/换入将活动地图保持在设备内存中。
• 实现了随机 Ferns 重定位器，以实现全局一致性的重定位和回环检测。