3D mapping using event cameras

光学相机已经非常成功地应用于纹理丰富的环境和良好的能见度条件下的三维视觉和机器人导航。然而，在环境非常动态或可视性较差的更复杂的场景中，它们有很强的局限性。在本文中，您将探索一种新的传感器模式，即事件相机（神经形态相机）以及它们如何帮助解决这些问题。事件相机提供非常高的帧率，可以在极端动态的环境中使用（每秒超过10k帧）。这些相机可以与主动/结构化照明技术一起使用，以重建3D轮廓，并在复杂和动态的环境中导航，例如水下领域，这将是该工作的主要应用重点。这个博士学位本质上是多学科的，在计算机视觉，机器人和计算成像之间的接口。该项目的目标是开发一个快速的3D地图系统和相关的数据处理管道，使用基于主动激光的照明和基于事件的（神经形态）相机，这是一种新型的生物启发相机。

Optical cameras have been very successfully used for 3D vision and robotic navigation in texture rich environments and good visibility conditions. However, they have strong limitations in more complex scenarios where the environment is either very dynamic or visibility is poor. In this thesis, you will explore a new sensor modality, namely event-cameras (neuromorphic cameras) and how they can help solve these problems. Event cameras offer very high frame rates and can be used in extremely dynamic environments (more than 10k frames per second). These cameras can be used with active/structured illumination techniques to reconstruct 3D profiles and to navigate in complex and dynamic environments such as the underwater domain which will be the primary application focus of the work. This PhD will be intrinsically multidisciplinary, at the interface between computer vision, robotics and computational imaging. The goal of this project is to develop a fast 3D mapping system and associated data processing pipeline using active laser-based illumination and event-based (neuromorphic) cameras, a new class of bio-inspired cameras.