EvaGaussians: Event Stream Assisted Gaussian Splatting from Blurry Images EvaGaussians is a novel method that integrates event streams from event cameras to assist in reconstructing high-quality 3D Gaussian Splatting (3D-GS) from blurry images. The method leverages the high temporal resolution and dynamic range of event cameras to explicitly model the formation process of motion-blurred images and guide the deblurring reconstruction of 3D-GS. By jointly optimizing the 3D-GS parameters and recovering camera motion trajectories during the exposure time, the method can robustly facilitate the acquisition of high-fidelity novel views with intricate texture details. The method was evaluated on a novel synthetic dataset and a newly collected real-world dataset. Both qualitative and quantitative comparisons demonstrate that EvaGaussians surpasses existing techniques in restoring fine details from blurry images and producing high-fidelity novel views. The method contributes two novel datasets, including a synthetic dataset with diverse scenes and a real-world dataset captured using a Color DAVIS346 event camera. The code and dataset will be released for future research.EvaGaussians: Event Stream Assisted Gaussian Splatting from Blurry Images EvaGaussians is a novel method that integrates event streams from event cameras to assist in reconstructing high-quality 3D Gaussian Splatting (3D-GS) from blurry images. The method leverages the high temporal resolution and dynamic range of event cameras to explicitly model the formation process of motion-blurred images and guide the deblurring reconstruction of 3D-GS. By jointly optimizing the 3D-GS parameters and recovering camera motion trajectories during the exposure time, the method can robustly facilitate the acquisition of high-fidelity novel views with intricate texture details. The method was evaluated on a novel synthetic dataset and a newly collected real-world dataset. Both qualitative and quantitative comparisons demonstrate that EvaGaussians surpasses existing techniques in restoring fine details from blurry images and producing high-fidelity novel views. The method contributes two novel datasets, including a synthetic dataset with diverse scenes and a real-world dataset captured using a Color DAVIS346 event camera. The code and dataset will be released for future research.