EaDeblur-GS: Event assisted 3D Deblur Reconstruction with Gaussian Splatting
This addresses limitations in 3D reconstruction for applications like robotics or AR/VR, though it appears incremental by building on existing 3DGS and event-based techniques.
The paper tackles the problem of 3D deblurring reconstruction from blurry images under severe blur and complex camera motion by integrating event camera data with Gaussian Splatting, achieving sharp 3D reconstructions in real-time with performance comparable to state-of-the-art methods.
3D deblurring reconstruction techniques have recently seen significant advancements with the development of Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS). Although these techniques can recover relatively clear 3D reconstructions from blurry image inputs, they still face limitations in handling severe blurring and complex camera motion. To address these issues, we propose Event-assisted 3D Deblur Reconstruction with Gaussian Splatting (EaDeblur-GS), which integrates event camera data to enhance the robustness of 3DGS against motion blur. By employing an Adaptive Deviation Estimator (ADE) network to estimate Gaussian center deviations and using novel loss functions, EaDeblur-GS achieves sharp 3D reconstructions in real-time, demonstrating performance comparable to state-of-the-art methods.