NeRF, introduced by Mildenhall et al. at ECCV 2020, is a 3D synthesis technique that represents a scene's radiance and density with a small MLP. Trained from a handful of photographs, the network can render photorealistic new views from previously unseen angles. It triggered a research explosion — variants like Instant-NGP cut training from hours to seconds. With Gaussian Splatting arriving in 2023 and largely displacing it for real-time rendering, NeRF nonetheless remains a core tool in digital-twin, robotics, and VFX pipelines.
MEVZU N°124ISTANBULYEAR I — VOL. III
Glossary · Advanced · 2020
NeRF — Neural Radiance Fields
An image-synthesis technique that represents a scene's 3D radiance field with a neural network.
- EN — English term
- NeRF
- TR — Turkish term
- NeRF — Sinirsel Radyans Alanı