Figure 3. The data generation process of event and RGB cameras (upper parts) and the pipeline of See-NeRF (lower parts) are shown in the
figure. We use NeRF to represent the actual radiance of an extreme lighting scene in the HDR domain. The volume rendering simulates the
raw scene radiance rays hitting the sensor and obtaining the raw sensor values. The predicted LDR images and events are generated with our
proposed sensor-physics grounded RGB and event CRF models (details are shown in the right part of the figure). The image loss and event
loss are employed to jointly supervise the optimization of the two CRF models and the NeRF network.