Filters








430 Hits in 5.4 sec

NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis [article]

Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, Ren Ng
2020 arXiv   pre-print
View synthesis results are best viewed as videos, so we urge readers to view our supplementary video for convincing comparisons.  ...  on neural rendering and view synthesis.  ...  Acknowledgements We thank Kevin Cao, Guowei Frank Yang, and Nithin Raghavan for comments and discussions. RR acknowledges funding from ONR grants N000141712687 and N000142012529 and the Ronald L.  ... 
arXiv:2003.08934v2 fatcat:2463k3mc35gvpg62v2jibcoyi4

NeRF

Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, Ren Ng
2022 Communications of the ACM  
on neural rendering and view synthesis.  ...  We describe how to effectively optimize neural radiance fields to render photorealistic novel views of scenes with complicated geometry and appearance, and demonstrate results that outperform prior work  ...  Acknowledgments We thank Kevin Cao, Guowei Frank Yang, and Nithin Raghavan for comments and discussions.  ... 
doi:10.1145/3503250 fatcat:uer4t5gk2fbo5ptv7y4ajawkq4

Recursive-NeRF: An Efficient and Dynamically Growing NeRF [article]

Guo-Wei Yang, Wen-Yang Zhou, Hao-Yang Peng, Dun Liang, Tai-Jiang Mu, Shi-Min Hu
2021 arXiv   pre-print
View synthesis methods using implicit continuous shape representations learned from a set of images, such as the Neural Radiance Field (NeRF) method, have gained increasing attention due to their high  ...  Now, an image of a scene can be rendered in a level-of-detail manner, so we posit that a complicated region of the scene should be represented by a large neural network while a small neural network is  ...  We would like to thank Guo-Ye Yang for his kindly help in experimentation and Prof. Ralph R. Martin for his help in writing.  ... 
arXiv:2105.09103v1 fatcat:zwm6shf425f5tc5nq5sxb5ywm4

Point-NeRF: Point-based Neural Radiance Fields [article]

Qiangeng Xu and Zexiang Xu and Julien Philip and Sai Bi and Zhixin Shu and Kalyan Sunkavalli and Ulrich Neumann
2022 arXiv   pre-print
Volumetric neural rendering methods like NeRF generate high-quality view synthesis results but are optimized per-scene leading to prohibitive reconstruction time.  ...  Point-NeRF combines the advantages of these two approaches by using neural 3D point clouds, with associated neural features, to model a radiance field.  ...  Neural radiance fields. NeRFs [35] have demonstrated remarkably high-quality results for novel view synthesis.  ... 
arXiv:2201.08845v5 fatcat:exwm66rh5zfzvooc4ouzbugmpi

NeRF-In: Free-Form NeRF Inpainting with RGB-D Priors [article]

Hao-Kang Liu, I-Chao Shen, Bing-Yu Chen
2022 arXiv   pre-print
Though Neural Radiance Field (NeRF) demonstrates compelling novel view synthesis results, it is still unintuitive to edit a pre-trained NeRF because the neural network's parameters and the scene geometry  ...  In this paper, we introduce the first framework that enables users to remove unwanted objects or retouch undesired regions in a 3D scene represented by a pre-trained NeRF without any category-specific  ...  Introduction Recent advancements in neural rendering, such as Neural Radiance Fields (NeRF) [23] has emerged as a powerful representation for the task of novel view synthesis, where the goal is to render  ... 
arXiv:2206.04901v1 fatcat:gcxxuknsijebplelvzddag5vry

HDR-NeRF: High Dynamic Range Neural Radiance Fields [article]

Xin Huang, Qi Zhang, Ying Feng, Hongdong Li, Xuan Wang, Qing Wang
2022 arXiv   pre-print
We present High Dynamic Range Neural Radiance Fields (HDR-NeRF) to recover an HDR radiance field from a set of low dynamic range (LDR) views with different exposures.  ...  Using the HDR-NeRF, we are able to generate both novel HDR views and novel LDR views under different exposures.  ...  Overall, HDR-NeRF can be represented by two continuous implicit neural functions: a radiance field for density and scene radiance and a tone mapper for color, as shown in Fig. 2 .  ... 
arXiv:2111.14451v3 fatcat:n446tyy4ebdvha6fv45zcndpe4

NeRF++: Analyzing and Improving Neural Radiance Fields [article]

Kai Zhang, Gernot Riegler, Noah Snavely, Vladlen Koltun
2020 arXiv   pre-print
Neural Radiance Fields (NeRF) achieve impressive view synthesis results for a variety of capture settings, including 360 capture of bounded scenes and forward-facing capture of bounded and unbounded scenes  ...  NeRF fits multi-layer perceptrons (MLPs) representing view-invariant opacity and view-dependent color volumes to a set of training images, and samples novel views based on volume rendering techniques.  ...  PRELIMINARIES Given posed multi-view images of a static scene, NeRF reconstructs an opacity field σ representing a soft shape, along with a radiance field c representing view-dependent surface texture.  ... 
arXiv:2010.07492v2 fatcat:jhrx4fu4ojfbpaugwxhkx2pkge

Continuous Dynamic-NeRF: Spline-NeRF [article]

Julian Knodt
2022 arXiv   pre-print
In order to demonstrate our architecture, we reconstruct dynamic scenes using Neural Radiance Fields, but hope it is clear that our approach is general and can be applied to a variety of problems.  ...  The problem of reconstructing continuous functions over time is important for problems such as reconstructing moving scenes, and interpolating between time steps.  ...  Static scene reconstruction is the problem of reconstructing a 3D scene from a set of 2D views of a nonchanging scene, and Neural Radiance Fields (NeRFs) [11] are a recent technique achieving this.  ... 
arXiv:2203.13800v1 fatcat:okdudn47frcbrfsgh4f3ulwcii

AD-NeRF: Audio Driven Neural Radiance Fields for Talking Head Synthesis [article]

Yudong Guo, Keyu Chen, Sen Liang, Yong-Jin Liu, Hujun Bao, Juyong Zhang
2021 arXiv   pre-print
Another advantage of our framework is that not only the head (with hair) region is synthesized as previous methods did, but also the upper body is generated via two individual neural radiance fields.  ...  In this paper, we address this problem with the aid of neural scene representation networks.  ...  Neural Radiance Fields for Talking Heads Based on the standard neural radiance field scene representation [30] and inspired by the dynamic neural radiance fields for facial animation introduced by Gafni  ... 
arXiv:2103.11078v3 fatcat:jcyv42gednfszjqxymkqojdaxe

Is Attention All NeRF Needs? [article]

Mukund Varma T, Peihao Wang, Xuxi Chen, Tianlong Chen, Subhashini Venugopalan, Zhangyang Wang
2022 arXiv   pre-print
We present Generalizable NeRF Transformer (GNT), a pure, unified transformer-based architecture that efficiently reconstructs Neural Radiance Fields (NeRFs) on the fly from source views.  ...  The first stage of GNT, called view transformer, leverages multi-view geometry as an inductive bias for attention-based scene representation, and predicts coordinate-aligned features by aggregating information  ...  Neural Radiance Field (NeRF) [40] achieves photorealistic and consistent novel view synthesis results by fitting each scene as a continuous 5D radiance field parameterized by an MLP.  ... 
arXiv:2207.13298v1 fatcat:f4ikucmkdvaq5boxq4x7u4ouoy

NeRF-Editing: Geometry Editing of Neural Radiance Fields [article]

Yu-Jie Yuan, Yang-Tian Sun, Yu-Kun Lai, Yuewen Ma, Rongfei Jia, Lin Gao
2022 arXiv   pre-print
Implicit neural rendering, especially Neural Radiance Field (NeRF), has shown great potential in novel view synthesis of a scene.  ...  However, current NeRF-based methods cannot enable users to perform user-controlled shape deformation in the scene.  ...  Our Method Our work is based on the neural radiance field (NeRF) [43] , which has promising performance in novel view synthesis.  ... 
arXiv:2205.04978v1 fatcat:kg5smbz75rg5vapyuwweql2b7y

D-NeRF: Neural Radiance Fields for Dynamic Scenes

Albert Pumarola, Enric Corona, Gerard Pons-Moll, Francesc Moreno-Noguer
2021 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Point of View & Time Figure 1: We propose D-NeRF, a method for synthesizing novel views, at an arbitrary point in time, of dynamic scenes with complex non-rigid geometries.  ...  The figure shows two scenes under variable points of view and time instances synthesised by the proposed model.  ...  Conclusion We have presented D-NeRF, a novel neural radiance field approach for modeling dynamic scenes.  ... 
doi:10.1109/cvpr46437.2021.01018 fatcat:qi3v4cwahfbmjdcesbvwiw7miq

Mega-NeRF: Scalable Construction of Large-Scale NeRFs for Virtual Fly-Throughs [article]

Haithem Turki, Deva Ramanan, Mahadev Satyanarayanan
2022 arXiv   pre-print
We use neural radiance fields (NeRFs) to build interactive 3D environments from large-scale visual captures spanning buildings or even multiple city blocks collected primarily from drones.  ...  To address these challenges, we begin by analyzing visibility statistics for large-scale scenes, motivating a sparse network structure where parameters are specialized to different regions of the scene  ...  Acknowledgments This research was supported by the National Science Foundation (NSF) under grant number CNS-2106862, the Defense Science and Technology Agency of Singapore (DSTA), and the CMU Argo AI Center for  ... 
arXiv:2112.10703v2 fatcat:zdiiudegzvewjhmc2unvhhiukm

NeRF–: Neural Radiance Fields Without Known Camera Parameters [article]

Zirui Wang, Shangzhe Wu, Weidi Xie, Min Chen, Victor Adrian Prisacariu
2022 arXiv   pre-print
Considering the problem of novel view synthesis (NVS) from only a set of 2D images, we simplify the training process of Neural Radiance Field (NeRF) on forward-facing scenes by removing the requirement  ...  synthesis quality as those trained with COLMAP pre-computed camera parameters.  ...  The authors would also like to thank Tim Yuqing Tang for insightful discussions and proofreading.  ... 
arXiv:2102.07064v4 fatcat:jzaoja2hxva2xnurnl5wqllcui

CLA-NeRF: Category-Level Articulated Neural Radiance Field [article]

Wei-Cheng Tseng, Hung-Ju Liao, Lin Yen-Chen, Min Sun
2022 arXiv   pre-print
We propose CLA-NeRF -- a Category-Level Articulated Neural Radiance Field that can perform view synthesis, part segmentation, and articulated pose estimation.  ...  During inference, it only takes a few RGB views (i.e., few-shot) of an unseen 3D object instance within the known category to infer the object part segmentation and the neural radiance field.  ...  Specifically, NeRF represents a scene as a volumetric field of density and RGB color .  ... 
arXiv:2202.00181v3 fatcat:dhfgzwezy5cefjxx3pkw3jgide
« Previous Showing results 1 — 15 out of 430 results