EVREAL

EVREAL数据集是由哈塞特佩大学和科克大学联合开发的一个综合评估和分析套件，专门用于基于事件的视频重建方法的研究。该数据集包含47700帧，涵盖了多种真实世界的测试序列，包括快速运动、低光和高动态范围场景。EVREAL数据集的创建旨在通过标准化评估协议和多样化的测试数据，促进不同方法之间的比较。此外，该数据集还支持对重建视频的质量进行定量和定性分析，以及对各种下游任务的性能评估，如相机校准、图像分类和物体检测。

The EVREAL dataset is a comprehensive evaluation and analysis suite jointly developed by Hacettepe University and University of Cork, specifically designed for research on event-based video reconstruction methods. This dataset contains 47,700 frames, covering a variety of real-world test sequences including fast motion, low-light, and high dynamic range (HDR) scenes. The EVREAL dataset was created to facilitate comparisons between different methods through standardized evaluation protocols and diverse test data. In addition, this dataset also supports both quantitative and qualitative analysis of reconstructed video quality, as well as performance evaluation for various downstream tasks such as camera calibration, image classification, and object detection.