MARF: The Medial Atom Ray Field Object Representation
Not on the map. Why?
Abstract (arXiv)
We propose Medial Atom Ray Fields (MARFs), a novel neural object representation that enables accurate differentiable surface rendering with a single network evaluation per camera ray. Existing neural ray fields struggle with multi-view consistency and representing surface discontinuities. MARFs address both using a medial shape representation, a dual representation of solid geometry that yields cheap geometrically grounded surface normals, in turn enabling computing analytical curvature despite the network having no second derivative. MARFs map a camera ray to multiple medial intersection candidates, subject to ray-sphere intersection testing. We illustrate how the learned medial shape quantities applies to sub-surface scattering, part segmentation, and aid representing a space of articulated shapes. Able to learn a space of shape priors, MARFs may prove useful for tasks like shape retrieval and shape completion, among others. Code and data can be found at https://github.com/pbsds/MARF.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 70 of its 103 references that are in the map. First posted on arXiv 30 Jun 2023.
Builds on
70 of 103 references.
References · 70
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
- 2016Layer Normalizationcited by 5,669
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
- 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
- 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
- 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
- 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
- 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,238
- 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,200
- 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,041
- 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,613
38 more
- 2018On the Spectral Bias of Neural Networkscited by 1,341
- 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,214
- 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
- 2021Volume Rendering of Neural Implicit Surfacescited by 996
- 2021PlenOctrees for Real-time Rendering of Neural Radiance Fieldscited by 905
- 2015Automatic differentiation in machine learning: a surveycited by 855
- 2020Accelerating 3D Deep Learning with PyTorch3Dcited by 784
- 2020Convolutional Occupancy Networkscited by 761
- 2020Implicit Geometric Regularization for Learning Shapescited by 710
- 2017Spectrally-normalized margin bounds for neural networkscited by 686
- 2021KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPscited by 673
- 2019Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelinescited by 672
- 2021UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstructioncited by 646
- 2021Neural Fields in Visual Computing and Beyondcited by 540
- 2018Meta-Learning with Latent Embedding Optimizationcited by 476
- 2021Baking Neural Radiance Fields for Real-Time View Synthesiscited by 468
- 2021Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapescited by 426
- 2019SAL: Sign Agnostic Learning of Shapes from Raw Datacited by 390
- 2020Local Implicit Grid Representations for 3D Scenescited by 373
- 2021Animatable Neural Radiance Fields for Modeling Dynamic Human Bodiescited by 361
- 2020Implicit Functions in Feature Space for 3D Shape Reconstruction and Completioncited by 343
- 2021NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Imagescited by 336
- 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 333
- 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 332
- 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
- 2020State of the Art on Neural Renderingcited by 278
- 2020Neural Unsigned Distance Fields for Implicit Function Learningcited by 265
- 2019Learning Shape Templates with Structured Implicit Functionscited by 242
- 2021DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networkscited by 232
- 2023MERF: Memory-Efficient Radiance Fields for Real-time View Synthesis in Unbounded Scenescited by 218
- 2020MetaSDF: Meta-learning Signed Distance Functionscited by 214
- 2021Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulationcited by 209
- 2020AutoInt: Automatic Integration for Fast Neural Volume Renderingcited by 204
- 2020DeRF: Decomposed Radiance Fieldscited by 163
- 2021Efficient Neural Radiance Fields for Interactive Free-viewpoint Videocited by 150
- 2021BACON: Band-limited Coordinate Networks for Multiscale Scene Representationcited by 143
- 2022Is Attention All That NeRF Needs?cited by 113
- 2021Animatable Neural Radiance Fields from Monocular RGB Videoscited by 109
50 most cited of 70 All 70 (JSON)
Built on by
Citing papers per year: 2025: 2.
2 papers.