Shape deformation for morphing or editing purposes is a central challenge in Computer Graphics. While numerous methods exist, few allow for the explicit evaluation of the deformation at arbitrary times and locations directly, without resorting to intricate advection or interpolation schemes. In this paper, we propose a method that provides an explicit expression of the deformation parameterized as a flow, for continuously deforming shapes defined implicitly. Implicit surfaces are indeed particularly well suited for deformation tasks, since they inherently account for both the surface and the enclosed volume. Our approach leverages invertible neural networks to ensure theoretically that the deformation is a valid flow, while also providing differential quantities useful for geometric regularization. We demonstrate applications of this flow to shape morphing with and without landmarks, shape editing, and pairwise-to-any morphing where we compute pairwise morphings to a canonical shape allowing to deduce transformations between any pair through flow composition. The code for our method is available at https://github.com/camillebnm/explicit_flows_for_implicit_surfaces.

This paper presents a novel topology-aware dimensionality reduction approach aiming at accurately visualizing the cyclic patterns present in high dimensional data. To that end, we build on the Topological Autoencoders (TopoAE) (Moor et al., 2020) formulation. First, we provide a novel theoretical analysis of its associated loss and show that a zero loss indeed induces identical persistence pairs (in high and low dimensions) for the 0-dimensional persistent homology ($\text{PH}^{0}$) of the Rips filtration. We also provide a counter example showing that this property no longer holds for a naive extension of TopoAE to $\text{PH}^{d}$ for $d\geq 1$. Based on this observation, we introduce a novel generalization of TopoAE to 1-dimensional persistent homology ($\text{PH}^{1}$), called TopoAE++, for the accurate generation of cycle-aware planar embeddings, addressing the above failure case. This generalization is based on the notion of cascade distortion, a new penalty term favoring an isometric embedding of the 2-chains filling persistent 1-cycles, hence resulting in more faithful geometrical reconstructions of the 1-cycles in the plane. We further introduce a novel, fast algorithm for the exact computation of $\text{PH}^{}$ for Rips filtrations in the plane, yielding improved runtimes over previously documented topology-aware methods. Our method also achieves a better balance between the topological accuracy, as measured by the Wasserstein distance, and the visual preservation of the cycles in low dimensions.

Abstract Shape interpolation is a long standing challenge of geometry processing. As it is ill‐posed, shape interpolation methods always work under some hypothesis such as semantic part matching or least displacement. Among such constraints, volume preservation is one of the traditional animation principles. In this paper we propose a method to interpolate between shapes in arbitrary poses favoring volume and topology preservation. To do so, we rely on a level set representation of the shape and its advection by a velocity field through the level set equation, both shape representation and velocity fields being parameterized as neural networks. While divergence free velocity fields ensure volume and topology preservation, they are incompatible with the Eikonal constraint of signed distance functions. This leads us to introduce the notion of adaptive divergence velocity field, a construction compatible with the Eikonal equation with theoretical guarantee on the shape volume preservation. In the non constant volume setting, our method is still helpful to provide a natural morphing, by combining it with a parameterization of the volume change over time. We show experimentally that our method exhibits better volume preservation than other recent approaches, limits topological changes and preserves the structures of shapes better without landmark correspondences.

Modern techniques for physical simulations rely on numerical schemes and mesh-refinement methods to address trade-offs between precision and complexity, but these handcrafted solutions are tedious and require high computational power. Data-driven methods based on large-scale machine learning promise high adaptivity by integrating long-range dependencies more directly and efficiently. In this work, we focus on computational fluid dynamics and address the shortcomings of a large part of the literature, which are based on fixed support for computations and predictions in the form of regular or irregular grids. We propose a novel setup to perform predictions in a continuous spatial and temporal domain while being trained on sparse observations. We formulate the task as a double observation problem and propose a solution with two interlinked dynamical systems defined on, respectively, the sparse positions and the continuous domain, which allows to forecast and interpolate a solution from the initial condition. Our practical implementation involves recurrent GNNs and a spatio-temporal attention observer capable of interpolating the solution at arbitrary locations. Our model not only generalizes to new initial conditions (as standard auto-regressive models do) but also performs evaluation at arbitrary space and time locations. We evaluate on three standard datasets in fluid dynamics and compare to strong baselines, which are outperformed in classical settings and the extended new task requiring continuous predictions.

Quasi-Monte Carlo integration is at the core of rendering. This technique estimates the value of an integral by evaluating the integrand at well-chosen sample locations. These sample points are designed to cover the domain as uniformly as possible to achieve better convergence rates than purely random points. Deterministic low-discrepancy sequences have been shown to outperform many competitors by guaranteeing good uniformity as measured by the so-called discrepancy metric, and, indirectly, by an integer t value relating the number of points falling into each domain stratum with the stratum area (lower t is better). To achieve randomness, scrambling techniques produce multiple realizations preserving the t value, making the construction stochastic. Among them, Owen scrambling is a popular approach that recursively permutes intervals for each dimension. However, relying on permutation trees makes it incompatible with smooth optimization frameworks. We present a differentiable Owen scrambling that regularizes permutations. We show that it can effectively be used with automatic differentiation tools for optimizing low-discrepancy sequences to improve metrics such as optimal transport uniformity, integration error, designed power spectra or projective properties, while maintaining their initial t -value as guaranteed by Owen scrambling. In some rendering settings, we show that our optimized sequences improve the rendering error.

Much effort has been put into developing samplers with specific properties, such as producing blue noise, low-discrepancy, lattice or Poisson disk samples. These samplers can be slow if they rely on optimization processes, may rely on a wide range of numerical methods, are not always differentiable. The success of recent diffusion models for image generation suggests that these models could be appropriate for learning how to generate point sets from examples. However, their convolutional nature makes these methods impractical for dealing with scattered data such as point sets. We propose a generic way to produce 2-d point sets imitating existing samplers from observed point sets using a diffusion model. We address the problem of convolutional layers by leveraging neighborhood information from an optimal transport matching to a uniform grid, that allows us to benefit from fast convolutions on grids, and to support the example-based learning of non-uniform sampling patterns. We demonstrate how the differentiability of our approach can be used to optimize point sets to enforce properties.

Abstract Optimal transport is a long‐standing theory that has been studied in depth from both theoretical and numerical point of views. Starting from the 50s this theory has also found a lot of applications in operational research. Over the last 30 years it has spread to computer vision and computer graphics and is now becoming hard to ignore. Still, its mathematical complexity can make it difficult to comprehend, and as such, computer vision and computer graphics researchers may find it hard to follow recent developments in their field related to optimal transport. This survey first briefly introduces the theory of optimal transport in layman's terms as well as most common numerical techniques to solve it. More importantly, it presents applications of these numerical techniques to solve various computer graphics and vision related problems. This involves applications ranging from image processing, geometry processing, rendering, fluid simulation, to computational optics, and many more. It is aimed at computer graphics researchers desiring to follow optimal transport research in their field as well as optimal transport researchers willing to find applications for their numerical algorithms.

Abstract Digital terrains are a foundational element in the computer‐generated depiction of natural scenes. Given the variety and complexity of real‐world landforms, there is a need for authoring solutions that achieve perceptually realistic outcomes without sacrificing artistic control. In this paper, we propose setting aside the elevation domain in favour of modelling in the gradient domain. Such a slope‐based representation is height independent and allows a seamless blending of disparate landforms from procedural, simulation, and real‐world sources. For output, an elevation model can always be recovered using Poisson reconstruction, which can include Dirichlet conditions to constrain the elevation of points and curves. In terms of authoring our approach has numerous benefits. It provides artists with a complete toolbox, including: cut‐and‐paste operations that support warping as needed to fit the destination terrain, brushes to modify region characteristics, and sketching to provide point and curve constraints on both elevation and gradient. It is also a unifying representation that enables the inclusion of tools from the spectrum of existing procedural and simulation methods, such as painting localised high‐frequency noise or hydraulic erosion, without breaking the formalism. Finally, our constrained reconstruction is GPU optimized and executes in real‐time, which promotes productive cycles of iterative authoring.

3D point clouds constitute an emerging multimedia content, now used in a wide range of applications. The main drawback of this representation is the size of the data since typical point clouds may contain millions of points, usually associated with both geometry and color information. Consequently, a significant amount of work has been devoted to the efficient compression of this representation. Lossy compression leads to a degradation of the data and thus impacts the visual quality of the displayed content. In that context, predicting perceived visual quality computationally is essential for the optimization and evaluation of compression algorithms. In this paper, we introduce PCQM, a full-reference objective metric for visual quality assessment of 3D point clouds. The metric is an optimally-weighted linear combination of geometry-based and color-based features. We evaluate its performance on an open subjective dataset of colored point clouds compressed by several algorithms; the proposed quality assessment approach outperforms all previous metrics in terms of correlation with mean opinion scores.

Abstract 3D acquisition of archaeological artefacts has become an essential part of cultural heritage research for preservation or restoration purpose. Statues, in particular, have been at the center of many projects. In this paper, we introduce a way to improve the understanding of acquired statues representing real or imaginary creatures by registering a simple and pliable articulated model to the raw point set data. Our approach performs a Forward And bacKward Iterative Registration (FAKIR) which proceeds joint by joint, needing only a few iterations to converge. We are thus able to detect the pose and elementary anatomy of sculptures, with possibly non realistic body proportions. By adapting our simple skeleton, our method can work on animals and imaginary creatures.

Being able to duplicate published research results is an important process of conducting research whether to build upon these findings or to compare with them. This process is called "replicability" when using the original authors' artifacts (e.g., code), or "reproducibility" otherwise (e.g., re-implementing algorithms). Reproducibility and replicability of research results have gained a lot of interest recently with assessment studies being led in various fields, and they are often seen as a trigger for better result diffusion and transparency. In this work, we assess replicability in Computer Graphics, by evaluating whether the code is available and whether it works properly. As a proxy for this field we compiled, ran and analyzed 151 codes out of 374 papers from 2014, 2016 and 2018 SIGGRAPH conferences. This analysis shows a clear increase in the number of papers with available and operational research codes with a dependency on the subfields, and indicates a correlation between code replicability and citation count. We further provide an interactive tool to explore our results and evaluation data.

In this paper, we present PC-MSDM, an objective metric for visual quality assessment of 3D point clouds. This full-reference metric is based on local curvature statistics and can be viewed as an extension for point clouds of the MSDM metric suited for 3D meshes. We evaluate its performance on an open subjective dataset of point clouds compressed by octree pruning; results show that the proposed metric outperforms its counterparts in terms of correlation with mean opinion scores.

Time-of-flight point cloud acquisition systems have grown in precision and robustness over the past few years. However, even subtle motion can induce significant distortions due to the long acquisition time. In contrast, there exists sensors that produce depth maps at a higher frame rate, but they suffer from low resolution and accuracy. In this paper, we correct distortions produced by small motions in time-of-flight acquisitions and even output a corrected animated sequence by combining a slow but high-resolution time-of-flight LiDAR system and a fast but low-resolution consumer depth sensor. We cast the problem as a curve-to-volume registration, by seeing a LiDAR point cloud as a curve in a 4-dimensional spacetime and the captured low-resolution depth video as a 4-dimensional spacetime volume. Our approach starts by registering both captured sequences in 4D, in a coarse-to-fine approach. It then computes an optical flow between the low-resolution frames and finally transfers high-resolution details by advecting along the flow. We demonstrate the efficiency of our approach on both synthetic data, on which we can compute registration errors, and real data.

Abstract Detail enhancement is a well‐studied area of 3D rendering and image processing, which has few equivalents for 3D shape processing. To enhance details, one needs an efficient analysis tool to express the local surface dynamics. We introduce Wavejets, a new function basis for locally decomposing a shape expressed over the local tangent plane, by considering both angular oscillations of the surface around each point and a radial polynomial. We link the Wavejets coefficients to surface derivatives and give theoretical guarantees for their precision and stability with respect to an approximate tangent plane. The coefficients can be used for shape details amplification, to enhance, invert or distort them, by operating either on the surface point positions or on the normals. From a practical point of view, we derive an efficient way of estimating Wavejets on point sets and demonstrate experimentally the amplification results with respect to noise or basis truncation.

Abstract Three‐dimensional scanners provide a virtual representation of object surfaces at some given precision that depends on many factors such as the object material, the quality of the laser ray or the resolution of the camera. This precision may even vary over the surface, depending, for example, on the distance to the scanner which results in uneven and unstructured point sets, with an uncertainty on the coordinates. To enhance the quality of the scanner output, one usually resorts to local surface interpolation between measured points. However, object surfaces often exhibit interesting statistical features such as repetitive geometric textures. Building on this property, we propose a new approach for surface super‐resolution that detects repetitive patterns or self‐similarities and exploits them to improve the scan resolution by aggregating scattered measures. In contrast with other surface super‐resolution methods, our algorithm has two important advantages. First, when handling multiple scans, it does not rely on surface registration. Second, it is able to produce super‐resolution from even a single scan. These features are made possible by a new local shape description able to capture differential properties of order above 2. By comparing those descriptors, similarities are detected and used to generate a high‐resolution surface. Our results show a clear resolution gain over state‐of‐the‐art interpolation methods.

We propose a new shape analysis approach based on the non-local analysis of local shape variations. Our method relies on a novel description of shape variations, called Local Probing Field (LPF), which describes how a local probing operator transforms a pattern onto the shape. By carefully optimizing the position and orientation of each descriptor, we are able to capture shape similarities and gather them into a geometrically relevant dictionary over which the shape decomposes sparsely. This new representation permits to handle shapes with mixed intrinsic dimensionality (e.g., shapes containing both surfaces and curves) and to encode various shape features such as boundaries. Our shape representation has several potential applications; here we demonstrate its efficiency for shape resampling and point set denoising for both synthetic and real data.

Authoring virtual terrains presents a challenge and there is a strong need for authoring tools able to create realistic terrains with simple user-inputs and with high user control. We propose an example-based authoring pipeline that uses a set of terrain synthesizers dedicated to specific tasks. Each terrain synthesizer is a Conditional Generative Adversarial Network trained by using real-world terrains and their sketched counterparts. The training sets are built automatically with a view that the terrain synthesizers learn the generation from features that are easy to sketch. During the authoring process, the artist first creates a rough sketch of the main terrain features, such as rivers, valleys and ridges, and the algorithm automatically synthesizes a terrain corresponding to the sketch using the learned features of the training samples. Moreover, an erosion synthesizer can also generate terrain evolution by erosion at a very low computational cost. Our framework allows for an easy terrain authoring and provides a high level of realism for a minimum sketch cost. We show various examples of terrain synthesis created by experienced as well as inexperienced users who are able to design a vast variety of complex terrains in a very short time.

LiDAR (Light Detection And Ranging) acquisition is a widespread method for measuring urban scenes, be it a small town neighborhood or an entire city. It is even more interesting when this acquisition is coupled with a collection of pictures registered with the data, permitting to recover the color information of the points. Yet, this added color can be perturbed by shadows that are very dependent on the sun direction and weather conditions during the acquisition. In this paper, we focus on the problem of automatically detecting and correcting the shadows from the LiDAR data by exploiting both the images and the point set laser reflectance. Building on the observation that shadow boundaries are characterized by both a significant color change and a stable laser reflectance, we propose to first detect shadow boundaries in the point set and then segment ground shadows using graph cuts in the image. Finally using a simplified illumination model we correct the shadows directly on the colored point sets. This joint exploitation of both the laser point set and the images renders our approach robust and efficient, avoiding user interaction.

Abstract In this paper, we present a simple and efficient method to represent terrains as elevation functions built from linear combinations of landform features (atoms). These features can be extracted either from real world data‐sets or procedural primitives, or from any combination of multiple terrain models. Our approach consists in representing the elevation function as a sparse combination of primitives, a concept which we call Sparse Construction Tree, which blends the different landform features stored in a dictionary. The sparse representation allows us to represent complex terrains using combinations of atoms from a small dictionary, yielding a powerful and compact terrain representation and synthesis tool. Moreover, we present a method for automatically learning the dictionary and generating the Sparse Construction Tree model. We demonstrate the efficiency of our method in several applications: inverse procedural modeling of terrains, terrain amplification and synthesis from a coarse sketch.

Abstract Most surfaces, be it from a fine‐art artifact or a mechanical object, are characterized by a strong self‐similarity. This property finds its source in the natural structures of objects but also in the fabrication processes: regularity of the sculpting technique, or machine tool. In this paper, we propose to exploit the self‐similarity of the underlying shapes for compressing point cloud surfaces which can contain millions of points at a very high precision. Our approach locally resamples the point cloud in order to highlight the self‐similarity of the shape, while remaining consistent with the original shape and the scanner precision. It then uses this self‐similarity to create an ad hoc dictionary on which the local neighborhoods will be sparsely represented, thus allowing for a light‐weight representation of the total surface. We demonstrate the validity of our approach on several point clouds from fine‐arts and mechanical objects, as well as a urban scene. In addition, we show that our approach also achieves a filtering of noise whose magnitude is smaller than the scanner precision.

Abstract This paper develops a scale space strategy for orienting and meshing exactly and completely a raw point set. The scale space is based on the intrinsic heat equation, also called mean curvature motion (MCM). A simple iterative scheme implementing MCM directly on the raw point set is described, and a mathematical proof of its consistency with MCM is given. Points evolved by this MCM implementation can be trivially backtracked to their initial raw position. Therefore, both the orientation and mesh of the data point set obtained at a smooth scale can be transported back on the original. The gain in visual accuracy is demonstrated on archaeological objects by comparison with several state of the art meshing methods.

Abstract For each scanned object 3D triangulation laser scanners deliver multiple sweeps corresponding to multiple laser motions and orientations. The problem of aligning these scans has been well solved by using rigid and, more recently, non‐rigid transformations. Nevertheless, there are always residual local offsets between scans which forbid a direct merging of the scans, and force to some preliminary smoothing. Indeed, the tiling and aliasing effects due to the tiniest normal displacements of the scans can be dramatic. This paper proposes a general method to tackle this problem. The algorithm decomposes each scan into its high and low frequency components and fuses the low frequencies while keeping intact the high frequency content. It produces a mesh with the highest attainable resolution, having for vertices all raw data points of all scans. This exhaustive fusion of scans maintains the finest texture details. The method is illustrated on several high resolution scans of archeological objects.