Editing Rendering (computer graphics) (section)

=== Neural approximations and light fields ===
A more recent, experimental approach is description of scenes using [[neural radiance field|radiance fields]] which define the color, intensity, and direction of incoming light at each point in space. (This is conceptually similar to, but not identical to, the [[light field]] recorded by a [[Holography|hologram]].) For any useful resolution, the amount of data in a radiance field is so large that it is impractical to represent it directly as volumetric data, and an [[approximation]] function must be found. [[Deep learning|Neural networks]] are typically used to generate and evaluate these approximations, sometimes using video frames, or a collection of photographs of a scene taken at different angles, as "[[Training, validation, and test data sets#Training data set|training data]]".{{r|n=Schmid2023}}{{r|n=Mildenhall2020}}

Algorithms related to neural networks have recently been used to find approximations of a scene as [[Gaussian splatting|3D Gaussians]]. The resulting representation is similar to a [[point cloud]], except that it uses fuzzy, partially-transparent blobs of varying dimensions and orientations instead of points. As with [[neural radiance field]]s, these approximations are often generated from photographs or video frames.{{r|n=Kerbl2023}}