Event-Camera Dataset and Simulator
收藏资源简介:
Event-Camera Dataset and Simulator是由苏黎世大学机器人与感知组创建的综合数据集,旨在支持高速和高动态范围的机器人和计算机视觉研究。该数据集包含多种合成和真实环境下的数据,包括全局快门强度图像、异步事件流、惯性测量和精确的相机姿态数据。数据集的创建涉及使用Blender软件生成渲染图像,并通过时间插值技术模拟事件相机数据。该数据集适用于开发和测试新的姿态估计、视觉测距和SLAM算法,特别适用于处理事件相机的非传统输出。
The Event-Camera Dataset and Simulator is a comprehensive dataset developed by the Robotics and Perception Group at the University of Zurich, designed to support high-speed and high-dynamic-range robotics and computer vision research. This dataset includes data from both synthetic and real-world environments, encompassing global-shutter intensity images, asynchronous event streams, inertial measurements, and precise camera pose data. The construction of this dataset involves using Blender software to generate rendered images, and simulating event camera data via temporal interpolation techniques. This dataset is suitable for developing and testing novel pose estimation, visual odometry, and SLAM algorithms, and is particularly useful for processing the non-traditional outputs of event cameras.




