DualSDF: Semantic Shape Manipulation Using a Two-Level Representation

Hao, Zekun; Averbuch‐Elor, Hadar; Snavely, Noah; Belongie, Serge

doi:10.1109/cvpr42600.2020.00765

Cited by 93 publications

(66 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some methods learn to decompose the shape into local parts automatically, and represent the parts with CSG primitives [32], superquadrics [28] or 3D Gaussians [12,13]. Dualsdf [18] expresses shapes at two levels of granularity, one capturing fine details and the other representing an abstracted proxy shape. Another set of work seeks to divide the 3D space into local patches.…”

Section: Related Workmentioning

confidence: 99%

TaylorImNet for Fast 3D Shape Reconstruction Based on Implicit Surface Function

Xiao¹,

Xu²,

Gao³

2022

Preprint

View full text Add to dashboard Cite

Benefiting from the contiguous representation ability, deep implicit functions can extract the iso-surface of a shape at arbitrary resolution. However, utilizing the neural network with a large number of parameters as the implicit function prevents the generation speed of high-resolution topology because it needs to forward a large number of query points into the network. In this work, we propose TaylorImNet inspired by the Taylor series for implicit 3D shape representation. TaylorImNet exploits a set of discrete expansion points and corresponding Taylor series to model a contiguous implicit shape field. After the expansion points and corresponding coefficients are obtained, our model only needs to calculate the Taylor series to evaluate each point and the number of expansion points is independent of the generating resolution. Based on this representation, our TaylorImNet can achieve a significantly faster generation speed than other baselines. We evaluate our approach on reconstruction tasks from various types of input, and the experimental results demonstrate that our approach can get slightly better performance than existing state-of-the-art baselines while improving the inference speed with a large margin. Our code will be released in https://github.com/svip-lab/TaylorImNet.

show abstract

Section: Related Workmentioning

confidence: 99%

TaylorImNet for Fast 3D Shape Reconstruction Based on Implicit Surface Function

Xiao¹,

Xu²,

Gao³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…We need to constrain the generator such that the location of the isosurface is within the spherical projection layer, and ideally approximate to the desired shape. Methods using geometric initializations [3], meta-learning [47], and variational methods [20] have been proposed to stabilize training and accelerate convergence. We choose to regularize our generator with a variational autodecoder (VAD) [20] loss alongside our adversarial criterion.…”

Section: Overall Modelmentioning

confidence: 99%

“…Methods using geometric initializations [3], meta-learning [47], and variational methods [20] have been proposed to stabilize training and accelerate convergence. We choose to regularize our generator with a variational autodecoder (VAD) [20] loss alongside our adversarial criterion. This VAD-GAN setup preserves the generator+discriminator structure of a generative adversarial network (GAN) during training, only requiring the addition of a light-weight embedding layer to the model, while significantly outperforming the original VAD only objective.…”

Section: Overall Modelmentioning

confidence: 99%

“…We compare against four alternative models for implicit shape generation: two variants of Shape-GAN [27] which use a voxel and PointNet as discriminator, a latent-GAN trained on DeepSDF embeddings, and a VAD based SDF model which does not utilize our surface discriminator. We use the VAD layer implementation provided in [20]. For the latent-GAN, VAD, and our SurfGen model, we use the DeepSDF network as backbone.…”

Section: Shape Generationmentioning

confidence: 99%

See 1 more Smart Citation

SurfGen: Adversarial 3D Shape Synthesis with Explicit Surface Discriminators

Luo¹,

Li²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Recent advances in deep generative models have led to immense progress in 3D shape synthesis. While existing models are able to synthesize shapes represented as voxels, point-clouds, or implicit functions, these methods only indirectly enforce the plausibility of the final 3D shape surface.Here we present a 3D shape synthesis framework (SurfGen) that directly applies adversarial training to the object surface. Our approach uses a differentiable spherical projection layer to capture and represent the explicit zero isosurface of an implicit 3D generator as functions defined on the unit sphere. By processing the spherical representation of 3D object surfaces with a spherical CNN in an adversarial setting, our generator can better learn the statistics of natural shape surfaces. We evaluate our model on largescale shape datasets, and demonstrate that the end-to-end trained model is capable of generating high fidelity 3D shapes with diverse topology. Code is available at https: //github.com/aluo-x/NeuralRaycaster.

show abstract

“…There has been considerable interest in learningbased methods for signed distance field modelling. And this is lately especially true in the area of deep learning [2,9,3,4,10,16,17,8,11] 4 .…”

Section: Experiments On Learned Sdbsmentioning

confidence: 99%

Theoretical and empirical analysis of a fast algorithm for extracting polygons from signed distance bounds

Markuš¹

2021

Preprint

View full text Add to dashboard Cite

We investigate an asymptotically fast method for transforming signed distance bounds into polygon meshes. This is achieved by combining sphere tracing (also known as ray marching) and one of the traditional polygonization schemes (e.g., Marching cubes). Let us call this approach Gridhopping. We provide theoretical and experimental evidence that it is of the O(N 2 log N ) computational complexity for a polygonization grid with N 3 cells. The algorithm is tested on both a set of primitive shapes as well as signed distance fields generated from point clouds by machine learning. Given its speed, simplicity and portability, we argue that it could prove useful during the modelling stage as well as in shape compression for storage.

show abstract

DualSDF: Semantic Shape Manipulation Using a Two-Level Representation

Cited by 93 publications

References 33 publications

TaylorImNet for Fast 3D Shape Reconstruction Based on Implicit Surface Function

TaylorImNet for Fast 3D Shape Reconstruction Based on Implicit Surface Function

SurfGen: Adversarial 3D Shape Synthesis with Explicit Surface Discriminators

Theoretical and empirical analysis of a fast algorithm for extracting polygons from signed distance bounds

Contact Info

Product

Resources

About