LSE-NeRF: Learning Sensor Modeling Errors for Deblured Neural Radiance Fields with RGB-Event Stereo
This addresses blur artifacts in NeRF reconstructions for applications like robotics or AR/VR, but it is incremental as it builds on existing NeRF and event camera methods.
The paper tackles the problem of reconstructing clear Neural Radiance Fields (NeRF) under fast camera motions by using blurry RGB images and event camera data in a stereo setup, resulting in improved reconstructions as evaluated on their introduced dataset and EVIMOv2.
We present a method for reconstructing a clear Neural Radiance Field (NeRF) even with fast camera motions. To address blur artifacts, we leverage both (blurry) RGB images and event camera data captured in a binocular configuration. Importantly, when reconstructing our clear NeRF, we consider the camera modeling imperfections that arise from the simple pinhole camera model as learned embeddings for each camera measurement, and further learn a mapper that connects event camera measurements with RGB data. As no previous dataset exists for our binocular setting, we introduce an event camera dataset with captures from a 3D-printed stereo configuration between RGB and event cameras. Empirically, we evaluate our introduced dataset and EVIMOv2 and show that our method leads to improved reconstructions. Our code and dataset are available at https://github.com/ubc-vision/LSENeRF.