EventVOT
收藏资源简介:
EventVOT是首个大规模高分辨率事件相机视觉跟踪数据集,由安徽大学计算机科学与技术学院创建。该数据集包含1141个视频,覆盖多种目标类别如行人、车辆、无人机等,视频分辨率为1280×720。数据集通过Prophesee EVK4–HD事件相机收集,特别适合于纯事件基础的物体跟踪研究。EventVOT旨在解决现有低分辨率事件跟踪数据集的限制,推动高速度、低延迟的视觉跟踪技术发展,适用于自动驾驶、无人机摄影、智能视频监控等领域。
EventVOT is the first large-scale high-resolution event camera visual tracking dataset, developed by the School of Computer Science and Technology, Anhui University. This dataset comprises 1141 video sequences covering a wide range of object categories including pedestrians, vehicles, drones and others, with a resolution of 1280×720. Acquired using the Prophesee EVK4–HD event camera, it is ideally suited for pure event-based object tracking research. EventVOT is designed to address the limitations of existing low-resolution event tracking datasets, and advance the development of high-speed, low-latency visual tracking technologies, with applications in fields such as autonomous driving, drone photography, intelligent video surveillance and so on.

- 1Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline安徽大学计算机科学与技术学院 · 2023年



