VIDEO RENDERING METHOD AND SYSTEM IN WHICH STATIC NERF MODEL AND DYNAMIC NERF MODEL ARE FUSED

Provided in the present invention are a video rendering method and system in which a static NeRF model and a dynamic NeRF model are fused. The method comprises: selecting a plurality of sampling points on a sampling ray of a training video, and training a static NeRF; on the basis of a static volume density and static color information which are extracted by means of the trained static NeRF, training a dynamic NeRF in combination with time information, and obtaining a dynamic volume density and dynamic color information of the sampling points; adding the dynamic volume density and the static volume density together, and then inputting same into an activation function network layer, so as to obtain a final volume density; adding the dynamic color information and the static color information together, and then inputting same into the activation function network layer, so as to obtain final color information; inputting, into the static NeRF and the dynamic NeRF, parameters for use in rendering to be performed, s