IMU-Blur

IMU-Blur commenced our evaluation by randomly selecting 8350 clear images (aka. backgrounds) from existing image datasets~\cite{zhou2017places,quattoni2009recognizing}. By capturing IMU data during the motion blur induced by the RealSense D455i camera, we synthesized a dataset of 8350 blurred images accompanied by corresponding blur heat maps. Ultimately, this dataset, namely IMU-Blur, contains 6680 triplets for training and 1670 triplets for testing. Our IMU-Blur dataset has significant advantages over widely used datasets such as GoPro~\cite{nah2017deep} and RealBlur~\cite{rim2020real}, which contain limited scene variations. It includes images captured in 6680 environments, eliminating the interference of features present in repeated scenes for network learning. Unlike existing blurred datasets mainly recorded by high-speed cameras, IMU-Blur is more accessible to synthesize towards massive-produced and real-world motion blur.

IMU-Blur项目伊始，便从现有的图像数据集中随机选取了8350张清晰图像（亦称背景图像）cite{zhou2017places,quattoni2009recognizing}。通过采集由RealSense D455i相机引发的动态模糊过程中的IMU数据，我们合成了包含8350张模糊图像及其相应的模糊热图的数据库。最终，该数据库，即IMU-Blur，内含6680个训练三元组和1670个测试三元组。相较于广泛使用的如GoProcite{nah2017deep}和RealBlurcite{rim2020real}等数据集，IMU-Blur在场景变化多样性方面具有显著优势，后者所包含的场景变化有限。IMU-Blur包含了在6680种不同环境中所采集的图像，从而消除了重复场景中特征对网络学习产生的干扰。与主要由高速相机记录的现有模糊图像数据集不同，IMU-Blur在合成大规模生产及现实世界运动模糊方面更具可及性。