연세대 인공지능학회 YAI

YAI 9기 박준영님이 나의야이아카데미아팀에서 작성한 글입니다. NeRF : Representing Scene as Neural Radiance Fields for View Synthesis [Mildenhall, Ben, et al. "Nerf: Representing scenes as neural radiance fields for view synthesis." Communications of the ACM 65.1 (2021): 99-106.](https://dl.acm.org/doi/abs/10.1145/3503250) Abstract 이 논문은 input으로 한정된 수의 3D scene을 획득, 이를 활용하여 continous volumetric scene function에 입각한 여러 방향에서..

Instant Neural Graphics Primitives with Multiresolution Hash Encoding https://arxiv.org/abs/2201.05989 Instant Neural Graphics Primitives with a Multiresolution Hash Encoding Neural graphics primitives, parameterized by fully connected neural networks, can be costly to train and evaluate. We reduce this cost with a versatile new input encoding that permits the use of a smaller network without sa..

NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis **YAI 8기 김현진님이 비전 논문 심화팀에서 작성한 글입니다. European Conference on Computer Vision (ECCV), 2020 (Best Paper Honorable Mention) https://arxiv.org/pdf/2003.08934.pdf 3. Neural Radiance Field Scene Representation 이 논문에서는 연속적인 장면들을 5D 벡터 함수의 형태 (Radiance Field)로 나타낸다. Input은 위치 (x, y, z), 방향 (θ, ϕ) 이고, Radiance와 Radiance의 양을 조절하는 volume..