Figure 1. Events corresponding to a blurry LDR image contain
spatial difference (High Dynamic Range) and temporal difference
(High Temporal Resolution) information of the scene radiance,
making it an ideal data for enhancing HDR sharp 3D scene recon-
struction. See-NeRF models the physical imaging process with
proposed RGB and event CRF models to bridge the sensor-physics
discrepancy, leveraging events to infer scene dynamics and achiev-
ing sharper and better HDR novel view synthesis results compared
to the physical unaligned method EvHDR-NeRF [5].