3DOR: Eurographics Workshop on 3D Object Retrieval

Permanent URI for this community

https://diglib.eg.org/handle/10.2312/307

Browse

Now showing 1 - 11 of 11

3D Volume Matching for Mesh Animation of Moving Actors
(The Eurographics Association, 2014) Blache, Ludovic; Loscos, Celine; Nocent, Olivier; Lucas, Laurent; Benjamin Bustos and Hedi Tabia and Jean-Philippe Vandeborre and Remco Veltkamp
4D multiview reconstruction of moving actors has many applications in the entertainment industry and although studios providing such services become more accessible, efforts have to be done in order to improve the underlying technology to produce high-quality 4D contents. In this paper, we enable surface matching for an animated mesh sequence in order to introduce coherence in the data. The context is provided by an indoor multi-camera system which performs synchronized video captures from multiple viewpoints in a chroma key studio. Our input is given by a volumetric silhouette-based reconstruction algorithm that generates a visual hull at each frame of the video sequence. These 3D volumetric models differ from one frame to another, in terms of structure and topology, which makes them very difficult to use in post-production and 3D animation software solutions. Our goal is to transform this input sequence of independent 3D volumes into a single dynamic volumetric structure, directly usable in post-production. These volumes are then transformed into an animated mesh. Our approach is based on a motion estimation procedure. An unsigned distance function on the volumes is used as the main shape descriptor and a 3D surface matching algorithm minimizes the interference between unrelated surface regions. Experimental results, tested on our multiview datasets, show that our method outperforms approaches based on optical flow when considering robustness over several frames.
Computing Local Binary Patterns on Mesh Manifolds for 3D Texture Retrieval
(The Eurographics Association, 2015) Werghi, Naoufel; Tortorici, Claudio; Berretti, Stefano; Bimbo, Alberto Del; I. Pratikakis and M. Spagnuolo and T. Theoharis and L. Van Gool and R. Veltkamp
In this paper, we present and experiment a novel approach for retrieving 3D geometric texture patterns on 2D mesh-manifolds (i.e., surfaces in the 3D space) using local binary patterns (LBP) constructed on the mesh. The method is based on the recently proposed mesh-LBP framework [WBD15]. Compared to its depth-image counterpart, the mesh-LBP is distinguished by the following features: a) inherits the intrinsic advantages of mesh surface (e.g., preservation of the full geometry); b) does not require normalization; c) can accommodate partial matching. Experiments conducted with public 3D models with geometric texture showcase the superiority of the mesh-LBP descriptors in comparison with competitive methods.
Coulomb Shapes: Using Electrostatic Forces for Deformation-invariant Shape Representation
(The Eurographics Association, 2014) Boscaini, Davide; Girdziusas, Ramunas; Bronstein, Michael M.; Benjamin Bustos and Hedi Tabia and Jean-Philippe Vandeborre and Remco Veltkamp
Canonical shape analysis is a popular method in deformable shape matching, trying to bring the shape into a canonical form that undoes its non-rigid deformations, thus reducing the problem of non-rigid matching into a rigid one. The canonization can be performed by measuring geodesic distances between all pairs of points on the shape and embedding them into a Euclidean space by means of multidimensional scaling (MDS), which reduces the intrinsic isometries of the shape into the extrinsic (Euclidean) isometries of the embedding space. A notable drawback of MDS-based canonical forms is their sensitivity to topological noise: different shape connectivity can affect dramatically the geodesic distances, resulting in a global distortion of the canonical form. In this paper, we propose a different shape canonization approach based on a physical model of electrostatic repulsion.We minimize the Coulomb energy subject to the local distance constraints between adjacent shape vertices. Our model naturally handles topological noise, allowing to 'tear' the shape at points of strong repulsion. Furthermore, the problem is computationally efficient, as it lends itself to fast multipole methods. We show experimental results in which our method compares favorably to MDS-based canonical forms.
A Descriptor for Voxel Shapes Based on the Skeleton Cut Space
(The Eurographics Association, 2016) Feng, Cong; Jalba, Andrei C.; Telea, Alexandru C.; A. Ferreira and A. Giachetti and D. Giorgi
Two-dimensional medial axes and three-dimensional curve skeletons have been long used for shape retrieval tasks. In contrast, and despite their ability to fully capture shape geometry and topology, three-dimensional surface skeletons have seen much less usage in this context. We present here a framework for shape matching and retrieval based on such surface skeletons. To this end, we construct a space of cuts generated by the surface skeleton, which has desirable invariance properties with respect to shape size, rotation, translation, pose, and noise. Next, we extract a histogram-based descriptor from this cut space, and discuss three different metrics to compare such histograms for shape retrieval. We illustrate our proposal by showing our descriptor's effectiveness in shape retrieval using a known shape-database benchmark.
Directed Curvature Histograms for Robotic Grasping
(The Eurographics Association, 2017) Schulz, Rodrigo; Guerrero, Pablo; Bustos, Benjamin; Ioannis Pratikakis and Florent Dupont and Maks Ovsjanikov
Three-dimensional descriptors are a common tool nowadays, used in a wide range of tasks. Most of the descriptors that have been proposed in the literature focus on tasks such as object recognition and identification. This paper proposes a novel three-dimensional local descriptor, structured as a set of histograms of the curvature observed on the surface of the object in different directions. This descriptor is designed with a focus on the resolution of the robotic grasping problem, especially on the determination of the orientation required to grasp an object. We validate our proposal following a data-driven approach using grasping information and examples generated using the Gazebo simulator and a simulated PR2 robot. Experimental results show that the proposed descriptor is well suited for the grasping problem, exceeding the performance observed with recent descriptors.
An Edit Distance for Reeb Graphs
(The Eurographics Association, 2016) Bauer, Ulrich; Fabio, Barbara Di; Landi, Claudia; A. Ferreira and A. Giachetti and D. Giorgi
We consider the problem of assessing the similarity of 3D shapes using Reeb graphs from the standpoint of robustness under perturbations. For this purpose, 3D objects are viewed as spaces endowed with real-valued functions, while the similarity between the resulting Reeb graphs is addressed through a graph edit distance. The cases of smooth functions on manifolds and piecewise linear functions on polyhedra stand out as the most interesting ones. The main contribution of this paper is the introduction of a general edit distance suitable for comparing Reeb graphs in these settings. This edit distance promises to be useful for applications in 3D object retrieval because of its stability properties in the presence of noise.
An Experimental Shape Matching Approach for Protein Docking
(The Eurographics Association, 2016) Fernandes, Francisco; Ferreira, Alfredo; A. Ferreira and A. Giachetti and D. Giorgi
Proteins play a vital role in biological processes, with their function being largely determined by their structure. It is important to know what a protein binds, where it binds, how it binds, and what is its final conformation. Several methodologies have been applied to solve this complex protein-protein docking problem, but the number of degrees of freedom renders this a very slow and computationally heavy challenge. To handle this problem, we propose a multi-level space partition approach to describe the three-dimensional shape of the protein. By combining two proteins in the same data structure we are able to easily detect the shape-complementary regions. Moreover, by directly integrating bio-energetic information, we can drive the algorithm by both parameters and provide a fast and efficient way to overcome some of the limitations of previous approaches.
Geometric Histograms of 3D Keypoints for Face Identification with Missing Parts
(The Eurographics Association, 2013) Berretti, Stefano; Werghi, Naoufel; Bimbo, Alberto del; Pala, Pietro; Umberto Castellani and Tobias Schreck and Silvia Biasotti and Ioannis Pratikakis and Afzal Godil and Remco Veltkamp
In this work, an original solution to 3D face identification is proposed, which supports recognition also in the case of probes with missing parts. Distinguishing traits of the face are captured by first extracting 3D keypoints of a face scan, then measuring how the face surface changes in the keypoints neighborhood using a local descriptor. To this end, an adaptation of the meshDOG algorithm to the case of 3D faces is proposed, together with a multi-ring geometric histogram descriptor. Face similarity is then evaluated by comparing local keypoint descriptors across inlier pairs of matching keypoints between probe and gallery scans. Experiments have been performed to assess the keypoints distribution and repeatability. Recognition accuracy of the proposed approach has been evaluated on the Bosphorus database, showing competitive results with respect to existing 3D face biometrics solutions.
LightNet: A Lightweight 3D Convolutional Neural Network for Real-Time 3D Object Recognition
(The Eurographics Association, 2017) Zhi, Shuaifeng; Liu, Yongxiang; Li, Xiang; Guo, Yulan; Ioannis Pratikakis and Florent Dupont and Maks Ovsjanikov
With the rapid growth of 3D data, accurate and efficient 3D object recognition becomes a major problem. Machine learning methods have achieved the state-of-the-art performance in the area, especially for deep convolutional neural networks. However, existing network models have high computational cost and are unsuitable for real-time 3D object recognition applications. In this paper, we propose LightNet, a lightweight 3D convolutional neural network for real-time 3D object recognition. It achieves comparable accuracy to the state-of-the-art methods with a single model and extremely low computational cost. Experiments have been conducted on the ModelNet and Sydney Urban Objects datasets. It is shown that our model improves the VoxNet model by relative 17.4% and 23.1% on the ModelNet10 and ModelNet40 benchmarks with less than 67% of training parameters. It is also demonstrated that the model can be applied in real-time scenarios.
Symmetry and Fourier Descriptor: A Hybrid Feature For NURBS based B-Rep Models Retrieval
(The Eurographics Association, 2014) Dang, Quoc Viet; Morin, Geraldine; Mouysset, Sandrine; Benjamin Bustos and Hedi Tabia and Jean-Philippe Vandeborre and Remco Veltkamp
As the number of models in 3D databases grows, an efficient 3D models indexing mechanism and a similarity measure to ease model retrieval are necessary. In this paper, we present a query-by-model framework for NURBS based B-Rep models retrieval that combines partial symmetry of the object and the Fourier shape descriptor of canonical 2D projections of the 3D models. In fact, most objects are composed by similar parts up to an isometry. By detecting the dominant partial symmetry of a given NURBS based B-Rep model, we define two canonical planes from which the Fourier descriptors are extracted to measure the similarity among 3D models.
Towards an Observer-oriented Theory of Shape Comparison
(The Eurographics Association, 2016) Frosini, Patrizio; A. Ferreira and A. Giachetti and D. Giorgi
In this position paper we suggest a possible metric approach to shape comparison that is based on a mathematical formalization of the concept of observer, seen as a collection of suitable operators acting on a metric space of functions. These functions represent the set of data that are accessible to the observer, while the operators describe the way the observer elaborates the data and enclose the invariance that he/she associates with them. We expose this model and illustrate some theoretical reasons that justify its possible use for shape comparison.

Browse

Browsing 3DOR: Eurographics Workshop on 3D Object Retrieval by Subject "and object representations"

Results Per Page

Sort Options