3D Keypoint Estimation Using Implicit Representation Learning

In this paper, we tackle the challenging problem of 3D keypoint estimation of general objects using a novel implicit representation. Previous works have demonstrated promising results for keypoint prediction through direct coordinate regression or heatmap-based inference. However, these methods are commonly studied for specific subjects, such as human bodies and faces, which possess fixed keypoint structures. They also suffer in several practical scenarios where explicit or complete geometry is not given, including images and partial point clouds. Inspired by the recent success of advanced implicit representation in reconstruction tasks, we explore the idea of using an implicit field to represent keypoints. Specifically, our key idea is employing spheres to represent 3D keypoints, thereby enabling the learnability of the corresponding signed distance field. Explicit key-points can be extracted subsequently by our algorithm based on the Hough transform. Quantitative and qualitative evaluations also show the superiority of our representation in terms of prediction accuracy.     

Optimizing Object Decomposition to Reduce Visual Artifacts in 3D Printing

We propose a method for the automatic segmentation of 3D objects into parts which can be individually 3D printed and then reassembled by preserving the visual quality of the final object. Our technique focuses on minimizing the surface affected by supports, decomposing the object into multiple parts whose printing orientation is automatically chosen. The segmentation reduces the visual impact on the fabricated model producing non-planar cuts that adapt to the object shape. This is performed by solving an optimization problem that balances the effects of supports and cuts, while trying to place both in occluded regions of the object surface. To assess the practical impact of the solution, we show a number of segmented, 3D printed and reassembled objects.     

Scalable and Efficient Functional Map Computations on Dense Meshes

We propose a new scalable version of the functional map pipeline that allows to efficiently compute correspondences between potentially very dense meshes. Unlike existing approaches that process dense meshes by relying on ad-hoc mesh simplification, we establish an integrated end-to-end pipeline with theoretical approximation analysis. In particular, our method overcomes the computational burden of both computing the basis, as well the functional and pointwise correspondence computation by approximating the functional spaces and the functional map itself. Errors in the approximations are controlled by theoretical upper bounds assessing the range of applicability of our pipeline. With this construction in hand, we propose a scalable practical algorithm and demonstrate results on dense meshes, which approximate those obtained by standard functional map algorithms at the fraction of the computation time. Moreover, our approach outperforms the standard acceleration procedures by a large margin, leading to accurate results even in challenging cases.     

Automatic Design of Cable-Tensioned Glass Shells

We propose an optimization algorithm for the design of post-tensioned architectural shell structures, composed of triangular glass panels, in which glass has a load-bearing function. Due to its brittle nature, glass can fail when it is subject to tensile forces. Hence, we enrich the structure with a cable net, which is specifically designed to post-tension the shell, relieving the underlying glass structure from tension. We automatically derive an optimized cable layout, together with the appropriate pre-load of each cable. The method is driven by a physically based static analysis of the shell subject to its service load. We assess our approach by applying non-linear finite element analysis to several real-scale application scenarios. Such a method of cable tensioning produces glass shells that are optimized from the material usage viewpoint since they exploit the high compression strength of glass. As a result, they are lightweight and robust. Both aesthetic and static qualities are improved with respect to grid shell competitors.     

Learning Part Boundaries from 3D Point Clouds

We present a method that detects boundaries of parts in 3D shapes represented as point clouds. Our method is based on a graph convolutional network architecture that outputs a probability for a point to lie in an area that separates two or more parts in a 3D shape. Our boundary detector is quite generic: it can be trained to localize boundaries of semantic parts or geometric primitives commonly used in 3D modeling. Our experiments demonstrate that our method can extract more accurate boundaries that are closer to ground-truth ones compared to alternatives. We also demonstrate an application of our network to fine-grained semantic shape segmentation, where we also show improvements in terms of part labeling performance.     

Cross-Shape Attention for Part Segmentation of 3D Point Clouds

We present a deep learning method that propagates point-wise feature representations across shapes within a collection for the purpose of 3D shape segmentation. We propose a cross-shape attention mechanism to enable interactions between a shape's point-wise features and those of other shapes. The mechanism assesses both the degree of interaction between points and also mediates feature propagation across shapes, improving the accuracy and consistency of the resulting point-wise feature representations for shape segmentation. Our method also proposes a shape retrieval measure to select suitable shapes for cross-shape attention operations for each test shape. Our experiments demonstrate that our approach yields state-of-the-art results in the popular PartNet dataset.     

PointSkelCNN: Deep Learning-Based 3D Human Skeleton Extraction from Point Clouds

A 3D human skeleton plays important roles in human shape reconstruction and human animation. Remarkable advances have been achieved recently in 3D human skeleton estimation from color and depth images via a powerful deep convolutional neural network. However, applying deep learning frameworks to 3D human skeleton extraction from point clouds remains challenging because of the sparsity of point clouds and the high nonlinearity of human skeleton regression. In this study, we develop a deep learning-based approach for 3D human skeleton extraction from point clouds. We convert 3D human skeleton extraction into offset vector regression and human body segmentation via deep learning-based point cloud contraction. Furthermore, a disambiguation strategy is adopted to improve the robustness of joint points regression. Experiments on the public human pose dataset UBC3V and the human point cloud skeleton dataset 3DHumanSkeleton compiled by the authors show that the proposed approach outperforms the state-of-the-art methods.     

Effective Annotations Over 3D Models

Annotation support in interactive systems is often considered a simple task by the CG community, since it entails the apparently easy selection of a region and its connection with some information. The reality appears more complex. The scope of this paper is two-fold: first, to review the status of this domain, discussing and characterizing several approaches proposed in literature to manage annotations over geometric models; second, to present in detail an innovative solution proposed and assessed in the framework of Cultural Heritage (CH) applications, called ClippingVolumes. At the annotation definition stage ClippingVolumes uses 3D data to characterize the annotation region; subsequently, annotations are visualized by adopting a two-pass rendering solution which uses stencil buffers, thus without introducing new geometric elements, changing the topology or duplicating geometry elements. It solves most of the issues that afflict the current state of the art, such as fragmentation, annotation transfer to multiple representations and multi-resolution data encoding. The latter is a mandatory requirement to produce efficient web-based systems. We implemented and we fully tested this approach in the framework of a complex system that supports the documentation of CH restoration projects.     

Attention And Positional Encoding Are (Almost) All You Need For Shape Matching

The fast development of novel approaches derived from the Transformers architecture has led to outstanding performance in different scenarios, from Natural Language Processing to Computer Vision. Recently, they achieved impressive results even in the challenging task of non-rigid shape matching. However, little is known about the capability of the Transformer-encoder architecture for the shape matching task, and its performances still remained largely unexplored. In this paper, we step back and investigate the contribution made by the Transformer-encoder architecture compared to its more recent alternatives, focusing on why and how it works on this specific task. Thanks to the versatility of our implementation, we can harness the bi-directional structure of the correspondence problem, making it more interpretable. Furthermore, we prove that positional encodings are essential for processing unordered point clouds. Through a comprehensive set of experiments, we find that attention and positional encoding are (almost) all you need for shape matching. The simple Transformer-encoder architecture, coupled with relative position encoding in the attention mechanism, is able to obtain strong improvements, reaching the current state-of-the-art.     

Interpolated corrected curvature measures for polygonal surfaces

A consistent and yet practically accurate definition of curvature onto polyhedral meshes remains an open problem. We propose a new framework to define curvature measures, based on the Corrected Normal Current, which generalizes the normal cycle: it uncouples the positional information of the polyhedral mesh from its geometric normal vector field, and the user can freely choose the corrected normal vector field at vertices for curvature computations. A smooth surface is then built in the Grassmannian ℝ³ × 𝕊² by simply interpolating the given normal vector field. Curvature measures are then computed using the usual Lipschitz–Killing forms, and we provide closed-form formulas per triangle. We prove a stability result with respect to perturbations of positions and normals. Our approach provides a natural scale-space for all curvature estimations, where the scale is given by the radius of the measuring ball. We show on experiments how this method outperforms state-of-the-art methods on clean and noisy data, and even achieves pointwise convergence on difficult polyhedral meshes like digital surfaces. The framework is also well suited to curvature computations using normal map information.     

Semi-Supervised 3D Shape Recognition via Multimodal Deep Co-training

3D shape recognition has been actively investigated in the field of computer graphics. With the rapid development of deep learning, various deep models have been introduced and achieved remarkable results. Most 3D shape recognition methods are supervised and learn only from the large amount of labeled shapes. However, it is expensive and time consuming to obtain such a large training set. In contrast to these methods, this paper studies a semi-supervised learning framework to train a deep model for 3D shape recognition by using both labeled and unlabeled shapes. Inspired by the co-training algorithm, our method iterates between model training and pseudo-label generation phases. In the model training phase, we train two deep networks based on the point cloud and multi-view representation simultaneously. In the pseudo-label generation phase, we generate the pseudo-labels of the unlabeled shapes using the joint prediction of two networks, which augments the labeled set for the next iteration. To extract more reliable consensus information from multiple representations, we propose an uncertainty-aware consistency loss function to combine the two networks into a multimodal network. This not only encourages the two networks to give similar predictions on the unlabeled set, but also eliminates the negative influence of the large performance gap between the two networks. Experiments on the benchmark ModelNet40 demonstrate that, with only 10% labeled training data, our approach achieves competitive performance to the results reported by supervised methods.     

An Occlusion-aware Edge-Based Method for Monocular 3D Object Tracking using Edge Confidence

We propose an edge-based method for 6DOF pose tracking of rigid objects using a monocular RGB camera. One of the critical problem for edge-based methods is to search the object contour points in the image corresponding to the known 3D model points. However, previous methods often produce false object contour points in case of cluttered backgrounds and partial occlusions. In this paper, we propose a novel edge-based 3D objects tracking method to tackle this problem. To search the object contour points, foreground and background clutter points are first filtered out using edge color cue, then object contour points are searched by maximizing their edge confidence which combines edge color and distance cues. Furthermore, the edge confidence is integrated into the edge-based energy function to reduce the influence of false contour points caused by cluttered backgrounds and partial occlusions. We also extend our method to multi-object tracking which can handle mutual occlusions. We compare our method with the recent state-of-art methods on challenging public datasets. Experiments demonstrate that our method improves robustness and accuracy against cluttered backgrounds and partial occlusions.     

State-of-the-art in Automatic 3D Reconstruction of Structured Indoor Environments

Creating high-level structured 3D models of real-world indoor scenes from captured data is a fundamental task which has important applications in many fields. Given the complexity and variability of interior environments and the need to cope with noisy and partial captured data, many open research problems remain, despite the substantial progress made in the past decade. In this survey, we provide an up-to-date integrative view of the field, bridging complementary views coming from computer graphics and computer vision. After providing a characterization of input sources, we define the structure of output models and the priors exploited to bridge the gap between imperfect sources and desired output. We then identify and discuss the main components of a structured reconstruction pipeline, and review how they are combined in scalable solutions working at the building level. We finally point out relevant research issues and analyze research trends.     

FAKIR: An algorithm for revealing the anatomy and pose of statues from raw point sets

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.     

A Variational Loop Shrinking Analogy for Handle and Tunnel Detection and Reeb Graph Construction on Surfaces

The humble loop shrinking property played a central role in the inception of modern topology but it has been eclipsed by more abstract algebraic formalisms. This is particularly true in the context of detecting relevant non-contractible loops on surfaces where elaborate homological and/or graph theoretical constructs are favored in algorithmic solutions. In this work, we devise a variational analogy to the loop shrinking property and show that it yields a simple, intuitive, yet powerful solution allowing a streamlined treatment of the problem of handle and tunnel loop detection. Our formalization tracks the evolution of a diffusion front randomly initiated on a single location on the surface. Capitalizing on a diffuse interface representation combined with a set of rules for concurrent front interactions, we develop a dynamic data structure for tracking the evolution on the surface encoded as a sparse matrix which serves for performing both diffusion numerics and loop detection and acts as the workhorse of our fully parallel implementation. The substantiated results suggest our approach outperforms state of the art and robustly copes with highly detailed geometric models. As a byproduct, our approach can be used to construct Reeb graphs by diffusion thus avoiding commonly encountered issues when using Morse functions.     

Topology-Aware Surface Reconstruction for Point Clouds

We present an approach to incorporate topological priors in the reconstruction of a surface from a point scan. We base the reconstruction on basis functions which are optimized to provide a good fit to the point scan while satisfying predefined topological constraints. We optimize the parameters of a model to obtain a likelihood function over the reconstruction domain. The topological constraints are captured by persistence diagrams which are incorporated within the optimization algorithm to promote the correct topology. The result is a novel topology-aware technique which can (i) weed out topological noise from point scans, and (ii) capture certain nuanced properties of the underlying shape which could otherwise be lost while performing surface reconstruction. We show results reconstructing shapes with multiple potential topologies, compare to other classical surface construction techniques, and show the completion of real scan data.     

DFR: Differentiable Function Rendering for Learning 3D Generation from Images

Learning-based 3D generation is a popular research field in computer graphics. Recently, some works adapted implicit function defined by a neural network to represent 3D objects and have become the current state-of-the-art. However, training the network requires precise ground truth 3D data and heavy pre-processing, which is unrealistic. To tackle this problem, we propose the DFR, a differentiable process for rendering implicit function representation of 3D objects into 2D images. Briefly, our method is to simulate the physical imaging process by casting multiple rays through the image plane to the function space, aggregating all information along with each ray, and performing a differentiable shading according to every ray's state. Some strategies are also proposed to optimize the rendering pipeline, making it efficient both in time and memory to support training a network. With DFR, we can perform many 3D modeling tasks with only 2D supervision. We conduct several experiments for various applications. The quantitative and qualitative evaluations both demonstrate the effectiveness of our method.     

Lightweight Curvature Estimation on Point Clouds with Randomized Corrected Curvature Measures

The estimation of differential quantities on oriented point cloud is a classical step for many geometry processing tasks in computer graphics and vision. Even if many solutions exist to estimate such quantities, they usually fail at satisfying both a stable estimation with theoretical guarantee, and the efficiency of the associated algorithm. Relying on the notion of corrected curvature measures [LRT22, LRTC20] designed for surfaces, the method introduced in this paper meets both requirements. Given a point of interest and a few nearest neighbours, our method estimates the whole curvature tensor information by generating random triangles within these neighbours and normalising the corrected curvature measures by the corrected area measure. We provide a stability theorem showing that our pointwise curvatures are accurate and convergent, provided the noise in position and normal information has a variance smaller than the radius of neighbourhood. Experiments and comparisons with the state-of-the-art confirm that our approach is more accurate and much faster than alternatives. The method is fully parallelizable, requires only one nearest neighbour request per point of computation, and is trivial to implement.     

Generating Texture for 3D Human Avatar from a Single Image using Sampling and Refinement Networks

There has been significant progress in generating an animatable 3D human avatar from a single image. However, recovering texture for the 3D human avatar from a single image has been relatively less addressed. Because the generated 3D human avatar reveals the occluded texture of the given image as it moves, it is critical to synthesize the occluded texture pattern that is unseen from the source image. To generate a plausible texture map for 3D human avatars, the occluded texture pattern needs to be synthesized with respect to the visible texture from the given image. Moreover, the generated texture should align with the surface of the target 3D mesh. In this paper, we propose a texture synthesis method for a 3D human avatar that incorporates geometry information. The proposed method consists of two convolutional networks for the sampling and refining process. The sampler network fills in the occluded regions of the source image and aligns the texture with the surface of the target 3D mesh using the geometry information. The sampled texture is further refined and adjusted by the refiner network. To maintain the clear details in the given image, both sampled and refined texture is blended to produce the final texture map. To effectively guide the sampler network to achieve its goal, we designed a curriculum learning scheme that starts from a simple sampling task and gradually progresses to the task where the alignment needs to be considered. We conducted experiments to show that our method outperforms previous methods qualitatively and quantitatively.