41-Issue 4

Permanent URI for this collection

https://diglib7.eg.org/handle/10.2312/2633205
Prague, Czech Republic & Virtual | July 4th - July 6th 2022
(Rendering - Symposium Papers track is available here.)
Table of Contents
Global Illumination
Path Guiding with Vertex Triplet Distributions
[full paper Image] [meta data Image]
Vincent Schüßler, Johannes Hanika, Alisa Jung, and Carsten Dachsbacher
Once-more Scattered Next Event Estimation for Volume Rendering
[full paper Image] [meta data Image]
Johannes Hanika, Andrea Weidlich, and Marc Droske
Temporally Sliced Photon Primitives for Time-of-flight Rendering
[full paper Image] [meta data Image]
Yang Liu, Shaojie Jiao, and Wojciech Jarosz
Sampling
Single-pass Stratified Importance Resampling
[full paper Image] [meta data Image]
Ege Ciklabakkal, Adrien Gruson, Iliyan Georgiev, Derek Nowrouzezahrai, and Toshiya Hachisuka
A Bidirectional Formulation for Walk on Spheres
[full paper Image] [meta data Image]
Yang Qi, Dario Seyb, Benedikt Bitterli, and Wojciech Jarosz
Volume Rendering
Automatic Feature Selection for Denoising Volumetric Renderings
[full paper Image] [meta data Image]
Xianyao Zhang, Melvin Ott, Marco Manzi, Markus Gross, and Marios Papas
BXDFs
A Microfacet-based Hair Scattering Model
[full paper Image] [meta data Image]
Weizhen Huang, Matthias B. Hullin, and Johannes Hanika
A Position-Free Path Integral for Homogeneous Slabs and Multiple Scattering on Smith Microfacets
[full paper Image] [meta data Image]
Benedikt Bitterli and Eugene d'Eon
Microsurface Transformations
[full paper Image] [meta data Image]
Asen Atanasov, Vladimir Koylazov, Rossen Dimov, and Alexander Wilkie
Material Modeling and Measurement
Controlling Material Appearance by Examples
[full paper Image] [meta data Image]
Yiwei Hu, Miloš Hašan, Paul Guerrero, Holly Rushmeier, and Valentin Deschaintre
Physics-Based Inverse Rendering using Combined Implicit and Explicit Geometries
[full paper Image] [meta data Image]
Guangyan Cai, Kai Yan, Zhao Dong, Ioannis Gkioulekas, and Shuang Zhao
Neural Rendering
Deep Flow Rendering: View Synthesis via Layer-aware Reflection Flow
[full paper Image] [meta data Image]
Pinxuan Dai and Ning Xie
Stylized Rendering
Recolorable Posterization of Volumetric Radiance Fields Using Visibility-Weighted Palette Extraction
[full paper Image] [meta data Image]
Kenji Tojo and Nobuyuki Umetani
pOp: Parameter Optimization of Differentiable Vector Patterns
[full paper Image] [meta data Image]
Marzia Riso, Davide Sforza, and Fabio Pellacini
Patterns and Noises
Point-Pattern Synthesis using Gabor and Random Filters
[full paper Image] [meta data Image]
Xingchang Huang, Pooran Memari, Hans-Peter Seidel, and Gurprit Singh

BibTeX (41-Issue 4)
                
@article{
10.1111:cgf.14597,

journal = {Computer Graphics Forum},
title = {{
Rendering 2022 CGF 41-4: Frontmatter}},

author = {
Ghosh, Abhijeet
 and 
Wei, Li-Yi
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14597}

}
                
@article{
10.1111:cgf.14582,

journal = {Computer Graphics Forum},
title = {{
Path Guiding with Vertex Triplet Distributions}},

author = {
Schüßler, Vincent
 and 
Hanika, Johannes
 and 
Jung, Alisa
 and 
Dachsbacher, Carsten
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14582}

}
                
@article{
10.1111:cgf.14583,

journal = {Computer Graphics Forum},
title = {{
Once-more Scattered Next Event Estimation for Volume Rendering}},

author = {
Hanika, Johannes
 and 
Weidlich, Andrea
 and 
Droske, Marc
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14583}

}
                
@article{
10.1111:cgf.14584,

journal = {Computer Graphics Forum},
title = {{
Temporally Sliced Photon Primitives for Time-of-flight Rendering}},

author = {
Liu, Yang
 and 
Jiao, Shaojie
 and 
Jarosz, Wojciech
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14584}

}
                
@article{
10.1111:cgf.14586,

journal = {Computer Graphics Forum},
title = {{
A Bidirectional Formulation for Walk on Spheres}},

author = {
Qi, Yang
 and 
Seyb, Dario
 and 
Bitterli, Benedikt
 and 
Jarosz, Wojciech
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14586}

}
                
@article{
10.1111:cgf.14585,

journal = {Computer Graphics Forum},
title = {{
Single-pass Stratified Importance Resampling}},

author = {
Ciklabakkal, Ege
 and 
Gruson, Adrien
 and 
Georgiev, Iliyan
 and 
Nowrouzezahrai, Derek
 and 
Hachisuka, Toshiya
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14585}

}
                
@article{
10.1111:cgf.14587,

journal = {Computer Graphics Forum},
title = {{
Automatic Feature Selection for Denoising Volumetric Renderings}},

author = {
Zhang, Xianyao
 and 
Ott, Melvin
 and 
Manzi, Marco
 and 
Gross, Markus
 and 
Papas, Marios
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14587}

}
                
@article{
10.1111:cgf.14588,

journal = {Computer Graphics Forum},
title = {{
A Microfacet-based Hair Scattering Model}},

author = {
Huang, Weizhen
 and 
Hullin, Matthias B.
 and 
Hanika, Johannes
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14588}

}
                
@article{
10.1111:cgf.14589,

journal = {Computer Graphics Forum},
title = {{
A Position-Free Path Integral for Homogeneous Slabs and Multiple Scattering on Smith Microfacets}},

author = {
Bitterli, Benedikt
 and 
d'Eon, Eugene
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14589}

}
                
@article{
10.1111:cgf.14590,

journal = {Computer Graphics Forum},
title = {{
Microsurface Transformations}},

author = {
Atanasov, Asen
 and 
Koylazov, Vladimir
 and 
Dimov, Rossen
 and 
Wilkie, Alexander
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14590}

}
                
@article{
10.1111:cgf.14591,

journal = {Computer Graphics Forum},
title = {{
Controlling Material Appearance by Examples}},

author = {
Hu, Yiwei
 and 
Hašan, Miloš
 and 
Guerrero, Paul
 and 
Rushmeier, Holly
 and 
Deschaintre, Valentin
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14591}

}
                
@article{
10.1111:cgf.14592,

journal = {Computer Graphics Forum},
title = {{
Physics-Based Inverse Rendering using Combined Implicit and Explicit Geometries}},

author = {
Cai, Guangyan
 and 
Yan, Kai
 and 
Dong, Zhao
 and 
Gkioulekas, Ioannis
 and 
Zhao, Shuang
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14592}

}
                
@article{
10.1111:cgf.14593,

journal = {Computer Graphics Forum},
title = {{
Deep Flow Rendering: View Synthesis via Layer-aware Reflection Flow}},

author = {
Dai, Pinxuan
 and 
Xie, Ning
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14593}

}
                
@article{
10.1111:cgf.14594,

journal = {Computer Graphics Forum},
title = {{
Recolorable Posterization of Volumetric Radiance Fields Using Visibility-Weighted Palette Extraction}},

author = {
Tojo, Kenji
 and 
Umetani, Nobuyuki
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14594}

}
                
@article{
10.1111:cgf.14595,

journal = {Computer Graphics Forum},
title = {{
pOp: Parameter Optimization of Differentiable Vector Patterns}},

author = {
Riso, Marzia
 and 
Sforza, Davide
 and 
Pellacini, Fabio
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14595}

}
                
@article{
10.1111:cgf.14596,

journal = {Computer Graphics Forum},
title = {{
Point-Pattern Synthesis using Gabor and Random Filters}},

author = {
Huang, Xingchang
 and 
Memari, Pooran
 and 
Seidel, Hans-Peter
 and 
Singh, Gurprit
},
year = {
2022},

publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},


ISSN = {1467-8659},

DOI = {
10.1111/cgf.14596}

}

Browse

Recent Submissions

Now showing 1 - 16 of 16
  • Thumbnail Image
    Item
    Rendering 2022 CGF 41-4: Frontmatter
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Ghosh, Abhijeet; Wei, Li-Yi; Ghosh, Abhijeet; Wei, Li-Yi
  • Thumbnail Image
    Item
    Path Guiding with Vertex Triplet Distributions
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Schüßler, Vincent; Hanika, Johannes; Jung, Alisa; Dachsbacher, Carsten; Ghosh, Abhijeet; Wei, Li-Yi
    Good importance sampling strategies are decisive for the quality and robustness of photorealistic image synthesis with Monte Carlo integration. Path guiding approaches use transport paths sampled by an existing base sampler to build and refine a guiding distribution. This distribution then guides subsequent paths in regions that are otherwise hard to sample. We observe that all terms in the measurement contribution function sampled during path construction depend on at most three consecutive path vertices. We thus propose to build a 9D guiding distribution over vertex triplets that adapts to the full measurement contribution with a 9D Gaussian mixture model (GMM). For incremental path sampling, we query the model for the last two vertices of a path prefix, resulting in a 3D conditional distribution with which we sample the next vertex along the path. To make this approach scalable, we partition the scene with an octree and learn a local GMM for each leaf separately. In a learning phase, we sample paths using the current guiding distribution and collect triplets of path vertices. We resample these triplets online and keep only a fixed-size subset in reservoirs. After each progression, we obtain new GMMs from triplet samples by an initial hard clustering followed by expectation maximization. Since we model 3D vertex positions, our guiding distribution naturally extends to participating media. In addition, the symmetry in the GMM allows us to query it for paths constructed by a light tracer. Therefore our method can guide both a path tracer and light tracer from a jointly learned guiding distribution.
  • Thumbnail Image
    Item
    Once-more Scattered Next Event Estimation for Volume Rendering
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Hanika, Johannes; Weidlich, Andrea; Droske, Marc; Ghosh, Abhijeet; Wei, Li-Yi
    We present a Monte Carlo path tracing technique to sample extended next event estimation contributions in participating media: we consider one additional scattering vertex on the way to the next event, accounting for focused blur, resulting in visually interesting image features. Our technique is tailored to thin homogeneous media with strongly forward scattering phase functions, such as water or atmospheric haze. Previous methods put emphasis on sampling transmittances or geometric factors, and are either limited to isotropic scattering, or used tabulation or polynomial approximation to account for some specific phase functions. We will show how to jointly importance sample the product of an arbitrary phase function with analytic sampling in the solid angle domain and the two reciprocal squared distance terms of the adjacent edges of the transport path. The technique is fast and simple to implement in an existing rendering system. Our estimator is designed specifically for forward scattering, so the new technique has to be combined with other estimators to cover the backward scattering contributions.
  • Thumbnail Image
    Item
    Temporally Sliced Photon Primitives for Time-of-flight Rendering
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Liu, Yang; Jiao, Shaojie; Jarosz, Wojciech; Ghosh, Abhijeet; Wei, Li-Yi
    We derive a class of new Monte Carlo estimators for volumetric time-of-flight rendering, generalizing prior work on transient photon points and beams. Conceptually, our method starts with any steady-state photon primitive – like a photon plane, parallelepiped, or parallelotope – and slices it with a temporal wavefront, producing a primitive of one dimension lower. We show how different unbiased temporally sliced primitives arise by analytically integrating any four dimensions within a novel extended spatio-temporal path space formulation. The differences between these estimators reduce to the determinant of a 4×4 Jacobian matrix, with columns dictated by the chosen dimensions. We then show how to combine the relative strengths of different sliced primitives using multiple importance sampling. Finally, we implement several of the new estimators enabled by our theory and compare them to each other as well as previous techniques.
  • Thumbnail Image
    Item
    A Bidirectional Formulation for Walk on Spheres
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Qi, Yang; Seyb, Dario; Bitterli, Benedikt; Jarosz, Wojciech; Ghosh, Abhijeet; Wei, Li-Yi
    Numerically solving partial differential equations (PDEs) is central to many applications in computer graphics and scientific modeling. Conventional methods for solving PDEs often need to discretize the space first, making them less efficient for complex geometry. Unlike conventional methods, the walk on spheres (WoS) algorithm recently introduced to graphics is a grid-free Monte Carlo method that can provide numerical solutions of Poisson equations without discretizing space. We draw analogies between WoS and classical rendering algorithms, and find that the WoS algorithm is conceptually equivalent to forward path tracing. Inspired by similar approaches in light transport, we propose a novel WoS reformulation that operates in the reverse direction, starting at source points and estimating the Green's function at ''sensor'' points. Implementations of this algorithm show improvement over classical WoS in solving Poisson equation with sparse sources. Our approach opens exciting avenues for future algorithms for PDE estimation which, analogous to light transport, connect WoS walks starting from sensors and sources and combine different strategies for robust solution algorithms in all cases.
  • Thumbnail Image
    Item
    Single-pass Stratified Importance Resampling
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Ciklabakkal, Ege; Gruson, Adrien; Georgiev, Iliyan; Nowrouzezahrai, Derek; Hachisuka, Toshiya; Ghosh, Abhijeet; Wei, Li-Yi
    Resampling is the process of selecting from a set of candidate samples to achieve a distribution (approximately) proportional to a desired target. Recent work has revisited its application to Monte Carlo integration, yielding powerful and practical importance sampling methods. One drawback of existing resampling methods is that they cannot generate stratified samples. We propose two complementary techniques to achieve efficient stratified resampling. We first introduce bidirectional CDF sampling which yields the same result as conventional inverse CDF sampling but in a single pass over the candidates, without needing to store them, similarly to reservoir sampling. We then order the candidates along a space-filling curve to ensure that stratified CDF sampling of candidate indices yields stratified samples in the integration domain. We showcase our method on various resampling-based rendering problems.
  • Thumbnail Image
    Item
    Automatic Feature Selection for Denoising Volumetric Renderings
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Zhang, Xianyao; Ott, Melvin; Manzi, Marco; Gross, Markus; Papas, Marios; Ghosh, Abhijeet; Wei, Li-Yi
    We propose a method for constructing feature sets that significantly improve the quality of neural denoisers for Monte Carlo renderings with volumetric content. Starting from a large set of hand-crafted features, we propose a feature selection process to identify significantly pruned near-optimal subsets. While a naive approach would require training and testing a separate denoiser for every possible feature combination, our selection process requires training of only a single probe denoiser for the selection task. Moreover, our approximate solution has an asymptotic complexity that is quadratic to the number of features compared to the exponential complexity of the naive approach, while also producing near-optimal solutions. We demonstrate the usefulness of our approach on various state-of-the-art denoising methods for volumetric content. We observe improvements in denoising quality when using our automatically selected feature sets over the hand-crafted sets proposed by the original methods.
  • Thumbnail Image
    Item
    A Microfacet-based Hair Scattering Model
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Huang, Weizhen; Hullin, Matthias B.; Hanika, Johannes; Ghosh, Abhijeet; Wei, Li-Yi
    The development of scattering models and rendering algorithms for human hair remains an important area of research in computer graphics. Virtually all available models for scattering off hair or fur fibers are based on separable lobes, which bring practical advantages in importance sampling, but do not represent physically-plausible microgeometry. In this paper, we contribute the first microfacet-based hair scattering model. Based on a rough cylinder geometry with tilted cuticle scales, our far-field model is non-separable by nature, yet allows accurate importance sampling. Additional benefits include support for elliptical hair cross-sections and an analytical solution for the reflected lobe using the GGX distribution. We show that our model captures glint-like forward scattering features in the R lobe that have been observed before but not properly explained.
  • Thumbnail Image
    Item
    A Position-Free Path Integral for Homogeneous Slabs and Multiple Scattering on Smith Microfacets
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Bitterli, Benedikt; d'Eon, Eugene; Ghosh, Abhijeet; Wei, Li-Yi
    We consider the problem of multiple scattering on Smith microfacets. This problem is equivalent to computing volumetric light transport in a homogeneous slab. Although the symmetry of the slab allows for significant simplification, fully analytic solutions are scarce and not general enough for most applications. Standard Monte Carlo simulation, although general, is expensive and leads to variance that must be dealt with. We present the first unbiased, truly position-free path integral for evaluating the BSDF of a homogeneous slab. We collapse the spatially-1D path integral of previous works to a position-free form using an analytical preintegration of collision distances. Evaluation of the resulting path integral, which now contains only directions, reduces to simple recursive manipulation of exponential distributions. Applying Monte Carlo to solve the reduced integration problem leads to lower variance. Our new algorithm allows us to render multiple scattering on Smith microfacets with less variance than prior work, and, in the case of conductors, to do so without any bias. Additionally, our algorithm can also be used to accelerate the rendering of BSDFs containing volumetrically scattering layers, at reduced variance compared to standard Monte Carlo integration.
  • Thumbnail Image
    Item
    Microsurface Transformations
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Atanasov, Asen; Koylazov, Vladimir; Dimov, Rossen; Wilkie, Alexander; Ghosh, Abhijeet; Wei, Li-Yi
    We derive a general result in microfacet theory: given an arbitrary microsurface defined via standard microfacet statistics, we show how to construct the statistics of its linearly transformed counterparts. A common use case of such transformations is to generate anisotropic versions of a given surface. Traditional anisotropic derivations based on varying the roughness of an isotropic distribution in an ellipse have a general closed-form formula only for the subclass of shape-invariant distributions. While our formulation is equivalent to these specific constructs, it is more general in two aspects: it leads to simple closedform solutions for all distributions, including shape-variant ones, and works for all invertible 2D transform matrices. The latter is of particular importance in case of deformation of the macrosurface, since it can be approximated locally by a linear transformation in the tangent plane. We demonstrate our results using the Generalized Trowbridge-Reitz (GTR) distribution which is shape-invariant only in the special case of the popular Trowbridge-Reitz (GGX) distribution.
  • Thumbnail Image
    Item
    Controlling Material Appearance by Examples
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Hu, Yiwei; Hašan, Miloš; Guerrero, Paul; Rushmeier, Holly; Deschaintre, Valentin; Ghosh, Abhijeet; Wei, Li-Yi
    Despite the ubiquitous use of materials maps in modern rendering pipelines, their editing and control remains a challenge. In this paper, we present an example-based material control method to augment input material maps based on user-provided material photos. We train a tileable version of MaterialGAN and leverage its material prior to guide the appearance transfer, optimizing its latent space using differentiable rendering. Our method transfers the micro and meso-structure textures of user provided target(s) photographs, while preserving the structure and quality of the input material. We show our methods can control existing material maps, increasing realism or generating new, visually appealing materials.
  • Thumbnail Image
    Item
    Physics-Based Inverse Rendering using Combined Implicit and Explicit Geometries
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Cai, Guangyan; Yan, Kai; Dong, Zhao; Gkioulekas, Ioannis; Zhao, Shuang; Ghosh, Abhijeet; Wei, Li-Yi
    Mathematically representing the shape of an object is a key ingredient for solving inverse rendering problems. 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. We introduce a new physics-based inverse rendering pipeline that uses both implicit and explicit representations. Our technique enjoys the benefit of both representations by supporting both topology changes and differentiable rendering of complex effects such as environmental illumination, soft shadows, and interreflection. We demonstrate the effectiveness of our technique using several synthetic and real examples.
  • Thumbnail Image
    Item
    Deep Flow Rendering: View Synthesis via Layer-aware Reflection Flow
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Dai, Pinxuan; Xie, Ning; Ghosh, Abhijeet; Wei, Li-Yi
    Novel view synthesis (NVS) generates images from unseen viewpoints based on a set of input images. It is a challenge because of inaccurate lighting optimization and geometry inference. Although current neural rendering methods have made significant progress, they still struggle to reconstruct global illumination effects like reflections and exhibit ambiguous blurs in highly viewdependent areas. This work addresses high-quality view synthesis to emphasize reflection on non-concave surfaces. We propose Deep Flow Rendering that optimizes direct and indirect lighting separately, leveraging texture mapping, appearance flow, and neural rendering. A learnable texture is used to predict view-independent features, meanwhile enabling efficient reflection extraction. To accurately fit view-dependent effects, we adopt a constrained neural flow to transfer image-space features from nearby views to the target view in an edge-preserving manner. Then we further implement a fusing renderer that utilizes the predictions of both layers to form the output image. The experiments demonstrate that our method outperforms the state-of-theart methods at synthesizing various scenes with challenging reflection effects.
  • Thumbnail Image
    Item
    Recolorable Posterization of Volumetric Radiance Fields Using Visibility-Weighted Palette Extraction
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Tojo, Kenji; Umetani, Nobuyuki; Ghosh, Abhijeet; Wei, Li-Yi
    Volumetric radiance fields have recently gained significant attention as promising representations of photorealistic scene reconstruction. However, the non-photorealistic rendering of such a representation has barely been explored. In this study, we investigate the artistic posterization of the volumetric radiance fields. We extend the recent palette-based image-editing framework, which naturally introduces intuitive color manipulation of the posterized results, into the radiance field. Our major challenge is applying stylization effects coherently across different views. Based on the observation that computing a palette frame-by-frame can produce flickering, we propose pre-computing a single palette from the volumetric radiance field covering its entire visible color. We present a method based on volumetric visibility to sample visible colors from the radiance field while avoiding occluded and noisy regions. We demonstrate our workflow by applying it to pre-trained volumetric radiance fields with various stylization effects. We also show that our approach can produce more coherent and robust stylization effects than baseline methods that compute a palette on each rendered view.
  • Thumbnail Image
    Item
    pOp: Parameter Optimization of Differentiable Vector Patterns
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Riso, Marzia; Sforza, Davide; Pellacini, Fabio; Ghosh, Abhijeet; Wei, Li-Yi
    Procedural materials are extensively used in computer graphics, since they provide editable, resolution-independent representation of textures. However, tuning the parameters of procedural generators to achieve a desired result remains time-consuming for users. Recently, inverse procedural material algorithms have been developed, exploiting differentiable rendering methods to find the parameters of a procedural model that match a target image. These approaches focus on raster textures. We propose pOp, a practical method for estimating the parameters of vector patterns, that are formed by collections of vector shapes arranged by an arbitrary procedural program. In our approach, patterns are defined as arbitrary programs, that control the translation, rotation and scale or vector graphics elements. We support elements typical of vector graphics, namely points, lines, circle, rounded rectangles, and quadratic Bèzier drawings, in multiple colors. We optimize the program parameters by automatically differentiating the signed distance field of the drawing, which we found to be significantly more reliable than using differentiable rendering of the final image. We demonstrate our method on a variety of cases, representing the variations found in structured vector patterns.
  • Thumbnail Image
    Item
    Point-Pattern Synthesis using Gabor and Random Filters
    (The Eurographics Association and John Wiley & Sons Ltd., 2022) Huang, Xingchang; Memari, Pooran; Seidel, Hans-Peter; Singh, Gurprit; Ghosh, Abhijeet; Wei, Li-Yi
    Point pattern synthesis requires capturing both local and non-local correlations from a given exemplar. Recent works employ deep hierarchical representations from VGG-19 [SZ15] convolutional network to capture the features for both point-pattern and texture synthesis. In this work, we develop a simplified optimization pipeline that uses more traditional Gabor transform-based features. These features when convolved with simple random filters gives highly expressive feature maps. The resulting framework requires significantly less feature maps compared to VGG-19-based methods [TLH19; RGF*20], better captures both the local and non-local structures, does not require any specific data set training and can easily extend to handle multi-class and multi-attribute point patterns, e.g., disk and other element distributions. To validate our pipeline, we perform qualitative and quantitative analysis on a large variety of point patterns to demonstrate the effectiveness of our approach. Finally, to better understand the impact of random filters, we include a spectral analysis using filters with different frequency bandwidths.