Computer Science

In light transport simulation, challenging situations are caused by the variety of materials and the relative length of path segments. Path Tracing can handle many situations and scales well to parallel hardware. However, it is not able to produce paths which have a smooth surface in connection with a small light source. Here, photon transports perform superior, which can be ineffective if the smooth object is small compared to the scene size. We propose to use the last segment of a Path Tracer path as the first segment of a photon path. As a result, the strengths of next event estimation are inherited by the photon transport and photons are guided toward the regions where they are most useful. To that end, we developed a lock-free sparse octree, which we use for fast and robust density estimates. Summarizing, the new method can outperform state of the art algorithms like Vertex Connection and Merging in certain scenarios.

In this technical report, we investigate efficient representations of articulated objects (e.g. human bodies), which is an important problem in computer vision and graphics. To deform articulated geometry, existing approaches represent objects as meshes and deform them using "skinning" techniques. The skinning operation allows a wide range of deformations to be achieved with a small number of control parameters. This paper introduces a method to invert the deformations undergone via traditional skinning techniques via a neural network parameterized by pose. The ability to invert these deformations allows values (e.g., distance function, signed distance function, occupancy) to be pre-computed at rest pose, and then efficiently queried when the character is deformed. We leave empirical evaluation of our approach to future work.

Point set registration is defined as a process to determine the spatial transformation from the source point set to the target one. Existing methods often iteratively search for the optimal geometric transformation to register a given pair of point sets, driven by minimizing a predefined alignment loss function. In contrast, the proposed point registration neural network (PR-Net) actively learns the registration pattern as a parametric function from a training dataset, consequently predict the desired geometric transformation to align a pair of point sets. PR-Net can transfer the learned knowledge (i.e. registration pattern) from registering training pairs to testing ones without additional iterative optimization. Specifically, in this paper, we develop novel techniques to learn shape descriptors from point sets that help formulate a clear correlation between source and target point sets. With the defined correlation, PR-Net tends to predict the transformation so that the source and target point sets can be statistically aligned, which in turn leads to an optimal spatial geometric registration. PR-Net achieves robust and superior performance for non-rigid registration of point sets, even in presence of Gaussian noise, outliers, and missing points, but requires much less time for registering large number of pairs. More importantly, for a new pair of point sets, PR-Net is able to directly predict the desired transformation using the learned model without repetitive iterative optimization routine. Our code is available at this https URL.

In a recent application of the Bokeh Python library for visualizing physico-chemical properties of chemical entities text-mined from the scientific literature, we found ourselves facing the task of smoothing hexagonally binned data in Cartesian coordinates. To the best of our knowledge, no documentation for how to do this exist in the public domain. This short paper shows how to accomplish this in general and for Bokeh in particular. We illustrate the method with a real-world example and discuss some potential advantages of using hexagonal bins in these and similar applications.

We introduce a novel computational framework for digital geometry processing, based upon the derivation of a nonlinear operator associated to the total variation functional. Such operator admits a generalized notion of spectral decomposition, yielding a sparse multiscale representation akin to Laplacian-based methods, while at the same time avoiding undesirable over-smoothing effects typical of such techniques. Our approach entails accurate, detail-preserving decomposition and manipulation of 3D shape geometry while taking an especially intuitive form: non-local semantic details are well separated into different bands, which can then be filtered and re-synthesized with a straightforward linear step. Our computational framework is flexible, can be applied to a variety of signals, and is easily adapted to different geometry representations, including triangle meshes and point clouds. We showcase our method throughout multiple applications in graphics, ranging from surface and signal denoising to detail transfer and cubic stylization.

Mesh denoising is a critical technology in geometry processing that aims to recover high-fidelity 3D mesh models of objects from their noise-corrupted versions. In this work, we propose a learning-based normal filtering scheme for mesh denoising called NormalNet, which maps the guided normal filtering (GNF) into a deep network. The scheme follows the iterative framework of filtering-based mesh denoising. During each iteration, first, the voxelization strategy is applied on each face in a mesh to transform the irregular local structure into the regular volumetric representation, therefore, both the structure and face normal information are preserved and the convolution operations in CNN(Convolutional Neural Network) can be easily performed. Second, instead of the guidance normal generation and the guided filtering in GNF, a deep CNN is designed, which takes the volumetric representation as input, and outputs the learned filtered normals. At last, the vertex positions are updated according to the filtered normals. Specifically, the iterative training framework is proposed, in which the generation of training data and the network training are alternately performed, whereas the ground truth normals are taken as the guidance normals in GNF to get the target normals. Compared to state-of-the-art works, NormalNet can effectively remove noise while preserving the original features and avoiding pseudo-features.

This paper presents and evaluates four weighting schemes for image interpolation algorithms. The first scheme is based on the normalized area of the circle, whose diameter is equal to the minimum side of a tetragon. The second scheme is based on the normalized area of the circle, whose radius is equal to the hypotenuse. The third scheme is based on the normalized area of the triangle, whose base and height are equal to the hypotenuse and virtual pixel length, respectively. The fourth weighting scheme is based on the normalized area of the circle, whose radius is equal to the virtual pixel length-based hypotenuse. Experiments demonstrated debatable algorithm performances and the need for further research.

Rendering large point clouds ordinarily requires building a hierarchical data structure for accessing the points that best represent the object for a given viewing frustum and level-of-detail. The building of such data structures frequently represents a large portion of the cost of the rendering pipeline both in terms of time and space complexity, especially when rendering is done for inspection purposes only. This problem has been addressed in the past by incremental construction approaches, but these either result in low quality hierarchies or in longer construction times. In this work we present OMiCroN -- Oblique Multipass Hierarchy Creation while Navigating -- which is the first algorithm capable of immediately displaying partial renders of the geometry, provided the cloud is made available sorted in Morton order. OMiCroN is fast, being capable of building the entire data structure in memory spending an amount of time that is comparable to that of just reading the cloud from disk. Thus, there is no need for storing an expensive hierarchy, nor for delaying the rendering until the whole hierarchy is read from disk. In fact, a pipeline coupling OMiCroN with an incremental sorting algorithm running in parallel can start rendering as soon as the first sorted prefix is produced, making this setup very convenient for streamed viewing.

We present the OpenAI Remote Rendering Backend (ORRB), a system that allows fast and customizable rendering of robotics environments. It is based on the Unity3d game engine and interfaces with the MuJoCo physics simulation library. ORRB was designed with visual domain randomization in mind. It is optimized for cloud deployment and high throughput operation. We are releasing it to the public under a liberal MIT license: this https URL .

We present a method for designing smooth cross fields on surfaces that automatically align to sharp features of an underlying geometry. Our approach introduces a novel class of energies based on a representation of cross fields in the spherical harmonic basis. We provide theoretical analysis of these energies in the smooth setting, showing that they penalize deviations from surface creases while otherwise promoting intrinsically smooth fields. We demonstrate the applicability of our method to quad-meshing and include an extensive benchmark comparing our fields to other automatic approaches for generating feature-aligned cross fields on triangle meshes.