Filters








11,313 Hits in 4.4 sec

Exploring Differential Geometry in Neural Implicits [article]

Tiago Novello, Guilherme Schardong, Luiz Schirmer, Vinicius da Silva, Helio Lopes, Luiz Velho
2022 arXiv   pre-print
We introduce a neural implicit framework that exploits the differentiable properties of neural networks and the discrete geometry of point-sampled surfaces to approximate them as the level sets of neural  ...  We also use the analytical derivatives of a neural implicit function to estimate the differential measures of the underlying point-sampled surface.  ...  Differentiable Neural Implicits This section explores the differential geometry of the level sets of networks during their training.  ... 
arXiv:2201.09263v4 fatcat:62u2ma2tmnakjaibzjhwx2qjye

A Level Set Theory for Neural Implicit Evolution under Explicit Flows [article]

Ishit Mehta, Manmohan Chandraker, Ravi Ramamoorthi
2022 arXiv   pre-print
Coordinate-based neural networks parameterizing implicit surfaces have emerged as efficient representations of geometry.  ...  geometry.  ...  Acknowledgements This work was supported in part by NSF CAREER 1751365, NSF IIS 2110409, ONR grant N000142012529, NSF Chase-CI grant 1730158, Adobe, Google, an Amazon Research Award, the Ronald L.  ... 
arXiv:2204.07159v2 fatcat:bubfu3vwpnf55no34cookkxt3m

Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance [article]

Lior Yariv, Yoni Kasten, Dror Moran, Meirav Galun, Matan Atzmon, Ronen Basri, Yaron Lipman
2020 arXiv   pre-print
We introduce a neural network architecture that simultaneously learns the unknown geometry, camera parameters, and a neural renderer that approximates the light reflected from the surface towards the camera  ...  The geometry is represented as a zero level-set of a neural network, while the neural renderer, derived from the rendering equation, is capable of (implicitly) modeling a wide set of lighting conditions  ...  Differentiable ray casting is mostly used with implicit shape representations such as implicit function defined over a volumetric grid or implicit neural representation, where the implicit function can  ... 
arXiv:2003.09852v3 fatcat:uix2ro3eevepnlnclz54og43t4

SelfRecon: Self Reconstruction Your Digital Avatar from Monocular Video [article]

Boyi Jiang, Yang Hong, Hujun Bao, Juyong Zhang
2022 arXiv   pre-print
We utilize differential mask loss of the explicit mesh to obtain the coherent overall shape, while the details on the implicit surface are refined with the differentiable neural rendering.  ...  Implicit representation supports arbitrary topology and can represent high-fidelity geometry shapes due to its continuous nature.  ...  For the implicit part, a differential formulation is designed to intersect the deformed surface and follow IDR's neural rendering [53] to refine the geometry.  ... 
arXiv:2201.12792v2 fatcat:ocubpi7ug5glxonubsl7crluhe

UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction [article]

Michael Oechsle, Songyou Peng, Andreas Geiger
2021 arXiv   pre-print
Neural implicit 3D representations have emerged as a powerful paradigm for reconstructing surfaces from multi-view images and synthesizing novel views.  ...  Our key insight is that implicit surface models and radiance fields can be formulated in a unified way, enabling both surface and volume rendering using the same model.  ...  We believe that neural implicit surfaces and advanced differentiable rendering procedures play a key role in future 3D reconstruction methods.  ... 
arXiv:2104.10078v2 fatcat:lp6qk36furafnj3tp2thrzoqty

Physics-Based Inverse Rendering using Combined Implicit and Explicit Geometries [article]

Guangyan Cai, Kai Yan, Zhao Dong, Ioannis Gkioulekas, Shuang Zhao
2022 arXiv   pre-print
Explicit representations like meshes are efficient to render in a differentiable fashion but have difficulties handling topology changes.  ...  Implicit representations like signed-distance functions, on the other hand, offer better support of topology changes but are much more difficult to use for physics-based differentiable rendering.  ...  Related Works In this paper, we address this problem by using both implicit and explicit representations for object geometry. Differentiable rendering of meshes.  ... 
arXiv:2205.01242v1 fatcat:342omlnqpbgmjl7ez5xmugb73u

DIST: Rendering Deep Implicit Signed Distance Function With Differentiable Sphere Tracing

Shaohui Liu, Yinda Zhang, Songyou Peng, Boxin Shi, Marc Pollefeys, Zhaopeng Cui
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Due to the nature of the implicit function, the rendering process requires tremendous function queries, which is particularly problematic when the function is represented as a neural network.  ...  We propose a differentiable sphere tracing algorithm to bridge the gap between inverse graphics methods and the recently proposed deep learning based implicit signed distance function.  ...  Differentiable Sphere Tracing In this section, we introduce our differentiable rendering method for the implicit signed distance function represented as a neural network, such as DeepSDF [34] .  ... 
doi:10.1109/cvpr42600.2020.00209 dblp:conf/cvpr/LiuZPSPC20 fatcat:3tvvsjfx3fbk7b3mzmlsfsdvae

Multiview Textured Mesh Recovery by Differentiable Rendering [article]

Lixiang Lin, Jianke Zhu, Yisu Zhang
2022 arXiv   pre-print
In contrast to the implicit neural representation on shape and color, we introduce a physically based inverse rendering scheme to jointly estimate the environment lighting and object's reflectance, which  ...  Specifically, a differentiable Poisson Solver is employed to represent the object's shape, which is able to produce topology-agnostic and watertight surfaces.  ...  Implicit Neural Representation Implicit neural representation directly estimates the objects' geometry from the input image through minimizing the photometric loss, which has achieved the encouraging results  ... 
arXiv:2205.12468v2 fatcat:y5kui3fbqzdn5owx7fsygugds4

Unified Implicit Neural Stylization [article]

Zhiwen Fan, Yifan Jiang, Peihao Wang, Xinyu Gong, Dejia Xu, Zhangyang Wang
2022 arXiv   pre-print
Our solution is a Unified Implicit Neural Stylization framework, dubbed INS.  ...  To regularize the geometry in 3D scenes, we propose a novel self-distillation geometry consistency loss which preserves the geometry fidelity of the stylized scenes.  ...  Neural Radiance Field: In contrast to point-wisely regression of implicit fields, NeRF [53] proposes to reconstruct a radiance field by inversing a differentiable rendering equation from captured images  ... 
arXiv:2204.01943v3 fatcat:sjluddeeznh2pbakb4kdfvwype

Development of a deep learning platform for optimising sheet stamping geometries subject to manufacturing constraints [article]

Hamid Reza Attar, Alistair Foster, Nan Li
2022 arXiv   pre-print
This approach features the interaction of two neural networks: 1) a geometry generator and 2) a manufacturing performance evaluator.  ...  These strategies enable the differentiable generation of high quality, large scale component geometries with tight local features for the first time.  ...  In contrast to these geometries, implicit neural representations they have not been used for modelling nor optimising stamping geometries before.  ... 
arXiv:2202.03422v1 fatcat:bszxi3hyczfono452dgnhovw5y

Critical Regularizations for Neural Surface Reconstruction in the Wild [article]

Jingyang Zhang, Yao Yao, Shiwei Li, Tian Fang, David McKinnon, Yanghai Tsin, Long Quan
2022 arXiv   pre-print
Neural implicit functions have recently shown promising results on surface reconstructions from multiple views.  ...  In this paper, we present RegSDF, which shows that proper point cloud supervisions and geometry regularizations are sufficient to produce high-quality and robust reconstruction results.  ...  Neural implicit surface reconstruction.  ... 
arXiv:2206.03087v1 fatcat:23tjyrtg4bfd7flss2jqwsigw4

Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering [article]

Vincent Sitzmann, Semon Rezchikov, William T. Freeman, Joshua B. Tenenbaum, Fredo Durand
2021 arXiv   pre-print
field parameterized via a neural implicit representation.  ...  Utilizing the analytical differentiability of neural implicit representations and a novel parameterization of light space, we further demonstrate the extraction of sparse depth maps from LFNs.  ...  field at any camera pose via analytical differentiation of the neural implicit representation.  ... 
arXiv:2106.02634v1 fatcat:niopcbm5cjhw5iwih4uv27oo54

Neural Surface Maps [article]

Luca Morreale, Noam Aigerman, Vladimir Kim, Niloy J. Mitra
2021 arXiv   pre-print
Maps are arguably one of the most fundamental concepts used to define and operate on manifold surfaces in differentiable geometry.  ...  Accordingly, in geometry processing, maps are ubiquitous and are used in many core applications, such as paramterization, shape analysis, remeshing, and deformation.  ...  [17] overfit neural networks to implicit fields of individual shapes, as a compact representation for their geometry.  ... 
arXiv:2103.16942v1 fatcat:dhalu4zjbbgc5p7xvzlzeq3oj4

On the Effectiveness of Weight-Encoded Neural Implicit 3D Shapes [article]

Thomas Davies and Derek Nowrouzezahrai and Alec Jacobson
2021 arXiv   pre-print
A neural implicit outputs a number indicating whether the given query point in space is inside, outside, or on a surface.  ...  In this paper, we establish that weight-encoded neural implicits meet the criteria of a first-class 3D shape representation.  ...  Our Neural Implicit can be smoothly interpolated in implicit space, and can be interactively modified with constructive solid geometry operations shown in Figure 10 .  ... 
arXiv:2009.09808v3 fatcat:7nvr4plxzbgevgqmtilk5zron4

DRaCoN – Differentiable Rasterization Conditioned Neural Radiance Fields for Articulated Avatars [article]

Amit Raj, Umar Iqbal, Koki Nagano, Sameh Khamis, Pavlo Molchanov, James Hays, Jan Kautz
2022 arXiv   pre-print
The output of DiffRas is then used as conditioning to our conditional neural 3D representation module (c-NeRF) which generates the final high-res image along with body geometry using volumetric rendering  ...  In this work, we present, DRaCoN, a framework for learning full-body volumetric avatars which exploits the advantages of both the 2D and 3D neural rendering techniques.  ...  In this work, we propose Differentiable Rasterization Conditioned Neural Radiance Field (DRaCoN), that leverages the advantages of both 2D and 3D neural rendering methods.  ... 
arXiv:2203.15798v1 fatcat:voygwvkn7rbh3pnsdghzv4447m
« Previous Showing results 1 — 15 out of 11,313 results