[Question / Feature Request] Comparisons with more recent NeRFs?
Opened this issue · 1 comments
xcyan commented
Great work! I am curious if the authors have ever compared the Deblur-NeRF with more recent NeRFs (e.g., MipNeRF, MipNeRF360) where motion blur is handled implicitly (by rendering conical frustums instead of individual rays)? Some clarifications would be preferred.
limacv commented
Thank you for your inspiring question. Rendering using conical frustums as in MipNeRF and MipNeRF tries to solve the problem of multi-scale rendering (or anti-aliasing) instead of modeling blurriness. So they inherently cannot handle motion blur. Actually, this multi-ray-based rendering used in DeblurNeRF can be easily integrated into the mipNeRF (i.e. rendering and combining multiple frustums). This could be an interesting potential future work : )