• Corpus ID: 244921004

Dense Depth Priors for Neural Radiance Fields from Sparse Input Views

  title={Dense Depth Priors for Neural Radiance Fields from Sparse Input Views},
  author={Barbara Roessle and Jonathan T. Barron and Ben Mildenhall and Pratul P. Srinivasan and Matthias Nie{\ss}ner},
Neural radiance fields (NeRF) encode a scene into a neural representation that enables photo-realistic rendering of novel views. However, a successful reconstruction from RGB images requires a large number of input views taken under static conditions — typically up to a few hundred images for room-size scenes. Our method aims to synthesize novel views of whole rooms from an order of magnitude fewer images. To this end, we leverage dense depth priors in order to constrain the NeRF optimization… 

Figures and Tables from this paper

NeRF-In: Free-Form NeRF Inpainting with RGB-D Priors
This paper introduces 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 data and training.
MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction
It is demonstrated that depth and normal cues, predicted by general-purpose monocular estimators, significantly improve reconstruction quality and optimization time, and geometric monocular priors improve performance both for small-scale single-object as well as large-scale multi-object scenes, independent of the choice of representation.
Decomposing NeRF for Editing via Feature Field Distillation
This work tackles the problem of semantic scene decomposition of NeRFs to enable query-based local editing of the represented 3D scenes, and distill the knowledge of off-the-shelf, self-supervised 2D image feature extractors into a 3D feature field optimized in parallel to the radiance field.
Advances in Neural Rendering
This state‐of‐the‐art report on advances in neural rendering focuses on methods that combine classical rendering principles with learned 3D scene representations, often now referred to as neural scene representations.
Sat-NeRF: Learning Multi-View Satellite Photogrammetry With Transient Objects and Shadow Modeling Using RPC Cameras
The Satellite Neural Radiance Field (Sat-NeRF), a new end-to-end model for learning multi-view satellite photogrammetry in the wild, is introduced and the advantages of applying a bundle adjustment to the satellite camera models prior to training are stressed.
RC-MVSNet: Unsupervised Multi-View Stereo with Neural Rendering
This work proposes a novel approach with neural rendering (RC-MVSNet) to solve ambiguity issues of correspondences among views of unsupervised Multi-View Stereo, imposing a depth rendering consistency loss to constrain the geometry features close to the object surface to alleviate occlusions.
NeRFReN: Neural Radiance Fields with Reflections
This work proposes to split a scene into transmitted and reflected components, and model the two components with separate neural radiance fields, and proposes to exploit geometric priors and apply carefully-designed training strategies to achieve reasonable decomposition results.
Advances in neural rendering
Loss functions for Neural Rendering Jun-Yan Zhu shows the importance of knowing the number of neurons in the system and how many neurons are firing at the same time.


Structure-from-Motion Revisited
This work proposes a new SfM technique that improves upon the state of the art to make a further step towards building a truly general-purpose pipeline.
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
This work describes how to effectively optimize neural radiance fields to render photorealistic novel views of scenes with complicated geometry and appearance, and demonstrates results that outperform prior work on neural rendering and view synthesis.
NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo
A new multi-view depth estimation method that utilizes both conventional SfM reconstruction and learning-based priors over the recently proposed neural radiance fields (NeRF), with surprising findings presented on the effectiveness of correspondence-based opti-mization and NeRF-based optimization over the adapted depth priors.
Depth-supervised NeRF: Fewer Views and Faster Training for Free
This work formalizes the above assumption through DS-NeRF (Depth-supervised Neural Radiance Fields), a loss for learning radiance that takes advantage of readily-available depth supervision and can render better images given fewer training views while training 2-3x faster.
Deep Residual Learning for Image Recognition
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
Free View Synthesis
This work presents a method for novel view synthesis from input images that are freely distributed around a scene that can synthesize images for free camera movement through the scene, and works for general scenes with unconstrained geometric layouts.
NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections
A learning-based method for synthesizing novel views of complex scenes using only unstructured collections of in-the-wild photographs, and applies it to internet photo collections of famous landmarks, to demonstrate temporally consistent novel view renderings that are significantly closer to photorealism than the prior state of the art.
SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scans
A novel approach that converts partial and noisy RGB-D scans into high-quality 3D scene reconstructions by inferring unobserved scene geometry and combined with a new 3D sparse generative convolutional neural network architecture is able to predict highly detailed surfaces in a coarse-to-fine hierarchical fashion.
DeepView: View Synthesis With Learned Gradient Descent
This work presents a novel approach to view synthesis using multiplane images (MPIs) that incorporates occlusion reasoning, improving performance on challenging scene features such as object boundaries, lighting reflections, thin structures, and scenes with high depth complexity.
Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelines
An algorithm for view synthesis from an irregular grid of sampled views that first expands each sampled view into a local light field via a multiplane image (MPI) scene representation, then renders novel views by blending adjacent local light fields.