Unverified50% confidenceFactExact time
NeRF (Neural Radiance Fields) is a 3D reconstruction method proposed in 2020 by a UC Berkeley team that uses a neural network to represent a continuous volumetric radiance field, taking spatial coordinates and viewing direction as input and outputting color and density.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedNeural Radiance Field (NeRF) technology was proposed in 2020, implicitly encoding the color and density of any position in space through a multilayer perceptron to reconstruct continuous three-dimensional scenes from sparse image collections.80% similarUnverifiedNeRF（Neural Radiance Field）于2020年提出，通过隐式神经表示编码三维场景，可高质量合成任意视角图像，但推理耗时达数十秒至数分钟78% similarUnverified神经辐射场（NeRF）技术于2020年提出，首次实现了从少量2D图像重建高质量3D场景的突破68% similarUnverifiedNVIDIA于2021年提出神经辐射缓存（Neural Radiance Caching，NRC），将场景稳定的辐射传输规律缓存进小型神经网络，光线反弹次数超过阈值时直接查询网络代替继续追踪66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/829532API
curl https://kongchang.com/api/v1/knowledge/claims/829532MCP
get_claim(id=829532)