Color Event Camera Dataset (CED)
收藏资源简介:
Color Event Camera Dataset (CED) 是由澳大利亚国立大学、苏黎世大学和莫纳什大学联合创建的首个彩色事件相机数据集,包含50分钟的彩色帧和事件数据。数据集涵盖了室内外多种场景,旨在推动基于事件的视觉研究。CED数据集通过Color-DAVIS346相机采集,该相机能够输出彩色事件和标准帧,分辨率为346x260像素。数据集创建过程中,使用了固定曝光和自动曝光技术以减少运动模糊,并提供了同步和时间戳的事件、原始图像和彩色图像。CED数据集适用于广泛的视觉研究领域,如图像重建、对象检测和光学流估计,旨在解决高动态范围、高时间分辨率和抗运动模糊等问题。
Color Event Camera Dataset (CED) is the first color event camera dataset jointly created by The Australian National University, University of Zurich and Monash University. It contains 50 minutes of color frame and event data. The dataset covers a variety of indoor and outdoor scenarios, aiming to advance event-based vision research. Collected via the Color-DAVIS346 camera, which can output both color events and standard frames with a resolution of 346×260 pixels, the CED dataset adopted both fixed exposure and auto-exposure techniques during its construction to reduce motion blur. Synchronized event data, raw images and color images with timestamps are provided. The CED dataset is applicable to a wide range of computer vision research fields, such as image reconstruction, object detection and optical flow estimation, and is designed to address key issues including high dynamic range, high temporal resolution and resistance to motion blur.




