3DOR 12

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/309

Browse

Now showing 1 - 19 of 19

SHREC'12 Track: Sketch-Based 3D Shape Retrieval
(The Eurographics Association, 2012) Li, B.; Schreck, Tobias; Godil, A.; Alexa, M.; Boubekeur, T.; Bustos, B.; Chen, J.; Eitz, M.; Furuya, T.; Hildebrand, K.; Huang, S.; Johan, H.; Kuijper, A.; Ohbuchi, R.; Richter, R.; Saavedra, J. M.; Scherer, M.; Yanagimachi, T.; Yoon, G. J.; Yoon, S. M.; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Sketch-based 3D shape retrieval has become an important research topic in content-based 3D object retrieval. The aim of this track is to measure and compare the performance of sketch-based 3D shape retrieval methods implemented by different participants over the world. The track is based on a new sketch-based 3D shape benchmark, which contains two types of sketch queries and two versions of target 3D models. In this track, 7 runs have been submitted by 5 groups and their retrieval accuracies were evaluated using 7 commonly used retrieval performance metrics. We hope that the benchmark, its corresponding evaluation code, and the comparative evaluation results of the state-of-the-art sketch-based 3D model retrieval algorithms will contribute to the progress of this research direction for the 3D model retrieval community.
SHREC'12 Track: Stability on Abstract Shapes
(The Eurographics Association, 2012) Biasotti, S.; Bai, X.; Bustos, B.; Cerri, A.; Giorgi, D.; Li, L.; Mortara, M.; Sipiran, I.; Zhang, S.; Spagnuolo, M.; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
This contribution reports the results of the SHREC 2012 track: Stability on Abstract Shapes. This track saw six registrations of which only three participants effectively sent the results of their runs.
Similarity Based Object Retrieval of Composite Neuronal Structures
(The Eurographics Association, 2012) Schulze, Florian; Trapp, Martin; Bühler, Katja; Lui, Tianxiao; Dickson, Barry; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Circuit Neuroscience tries to solve one of the most challenging questions in biology: How does the brain work? An important step towards an answer to this question is to gather detailed knowledge about the neuronal circuits of the model organism Drosophila melanogaster. Geometric representations of neuronal objects of the Drosophila are acquired using molecular genetic methods, confocal microscopy, non-rigid registration and segmentation. These objects are integrated into a constantly growing common atlas. The comparison of new segmented neurons to already known neurons is a frequent task which evolves with a growing amount of data into a bottleneck of the knowledge discovery process. Thus, the exploration of the atlas by means of domain specific similarity measures becomes a pressing need. To enable similarity based retrieval of neuronal objects we defined together with domain experts tailored dissimilarity measures for each of the three typical neuronal sub structures cell body, projection, arborization. The dissimilarity measure for composite neurons has been defined as domain specific combination of the sub structure dissimilarities. According to domain experts the developed system has big advantages for all tasks which involve extensive data exploration.
Key-component Detection on 3D Meshes using Local Features
(The Eurographics Association, 2012) Sipiran, Ivan; Bustos, Benjamin; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
In this paper, we present a method to detect stable components on 3D meshes. A component is a region on the mesh which contains discriminative local features. Our goal is to represent a 3D mesh with a set of regions, which we called key-components, that characterize the represented object and therefore, they could be used for effective matching and recognition. As key-components are features in coarse scales, they are less sensitive to mesh deformations such as noise. In addition, the number of key-components is low compared to other local representations such as keypoints, allowing us to use them in efficient subsequent tasks. An desirable characteristic of a decomposition is that the components should be repeatable regardless shape transformations. We show in the experiments that the key-components are repeatable under several transformations using the SHREC'2010 feature detection benchmark.
Articulated Motion Segmentation of Point Clouds by Group-Valued Regularization
(The Eurographics Association, 2012) Rosman, Guy; Bronstein, Alexander M.; Bronstein, Michael M.; Kimmel, Ron; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Motion segmentation for articulated objects is an important topic of research. Yet such a segmentation should be as free as possible from underlying assumptions so as to fit general scenes and objects. In this paper we demonstrate an algorithm for articulated motion segmentation of 3D point clouds, free of any assumptions on the underlying model and yet firmly set in a well-defined variational framework. Results on scanned images show the generality of the proposed technique and its robustness to scanning artifacts and noise.
Sketch-based 3D Model Retrieval using Keyshapes for Global and Local Representation
(The Eurographics Association, 2012) Saavedra, Jose M.; Bustos, Benjamin; Schreck, Tobias; Yoon, Sang Min; Scherer, Maximiliam; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Since 3D models are becoming more popular, the need for effective methods capable of retrieving 3D models is becoming crucial. Current methods require an example 3D model as query. However, in many cases, such a query is not easy to get. An alternative is using a hand-drawn sketch as query. In this work, we present a new keyshape based approach named HKO-KASD for retrieving 3D models using rough sketches as queries. Our approach comprises two general steps. First, a global descriptor is used to determine the appropriate viewpoint for each model. Second, we apply a local matching process to determine the final ranking for an input sketch. To this end, we present a local descriptor capable of working with sketch representations. The global descriptors as well as the local descriptors rely on a set of keyshapes precomputed from 2D representations of 3D models and from the query sketch as well. We evaluate our method using the first-tier precision and compare it with current approaches (HELO, STELA). Our results show a significant increase in precision for many classes of 3D models.
Automated Contextual Annotation of B-Rep Cad Mechanical Components Deriving Technology and Symmetry Information to Support Partial Retrieval
(The Eurographics Association, 2012) Li, K.; Shahwan, A.; Trlin, M.; Foucault, G.; Léon, Jean-Claude; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Mechanical components are often related to technological meaning: it is a screw, a ball bearing, . . . Technological meaning relates to the notion of function which, in turn, refers to the environment of a component. Here, it is shown how a digital mock-up contextualizing a set of components is used to annotate its components with technological information. This information is derived from the interfaces between components. Interfaces together with the concept of state of a digital mock-up initiate annotations of components with functional information. Symmetries of components is a complementary information that can contribute to a query. Local rotational symmetry is analyzed and adapted to the context of mechanical components. Then, combining these two categories of information enables the generation of high level queries addressing a subset of a component boundary. Technological as well as symmetry informations are component annotations automatically derived from the digital mock-up analysis and from components analyzed separately.
Intrinsic Local Symmetries: A Computational Framework
(The Eurographics Association, 2012) Grushko, Carmi; Raviv, Dan; Kimmel, Ron; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
We present a computational framework for finding metric-preserving tangent vector fields on surfaces, also known as Killing Vector Fields. Flows of such vector fields define self-isometries of the surface, or in other words, symmetries. Our approach is based on general-purpose isometry-finding frameworks, and is shown to be robust to noise. In addition, we demonstrate symmetry recovery using non-Euclidean metrics.
Real-time Expression Recognition from Dynamic Sequences of 3D Facial Scans
(The Eurographics Association, 2012) Berretti, Stefano; Bimbo, Alberto del; Pala, Pietro; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
In this paper, we address the problem of person-independent facial expression recognition in dynamic sequences of 3D face scans. To this end, an original approach is proposed that relies on automatically extracting a set of 3D facial points, and modeling their mutual distances along time. Training an Hidden Markov Model for every prototypical facial expression to be recognized, and combining them to form a multi-class classifier, an average recognition rate of 76.3% on the angry, happy and surprise expressions of the BU-4DFE database has been obtained. Comparison with competitor approaches on the same database shows that our solution is able to obtain effective results with the clear advantage of an implementation that fits to real-time constraints.
3D Object Retrieval via Range Image Queries based on SIFT descriptors on Panoramic Views
(The Eurographics Association, 2012) Sfikas, Konstantinos; Pratikakis, Ioannis; Theoharis, Theoharis; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
The increasing availability of low-cost 3D scanners is resulting in the creation of large repositories of 3D models. Low-cost 3D range scanners in particular can also be used to capture partial views of real 3D objects which can then act as queries over 3D object repositories. This paper concerns a new methodology for 3D object retrieval based on range image queries which represent partial views of 3D objects. SIFT descriptors based on panoramic views are used to address this problem. The proposed method is evaluated against state-of-the-art works on a standard dataset.
Affine-Invariant Photometric Heat Kernel Signatures
(The Eurographics Association, 2012) Kovnatsky, Artiom; Bronstein, Michael M.; Bronstein, Alexander M.; Raviv, Dan; Kimmel, Ron; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
In this paper, we explore the use of the diffusion geometry framework for the fusion of geometric and photometric information in local shape descriptors. Our construction is based on the definition of a modified metric, which combines geometric and photometric information, and then the diffusion process on the shape manifold is simulated. Experimental results show that such data fusion is useful in coping with shape retrieval experiments, where pure geometric and pure photometric methods fail. Apart from retrieval task the proposed diffusion process may be employed in other applications.
Spectral Analysis Driven Sparse Matching of 3D Shapes
(The Eurographics Association, 2012) Darom, Tal; Keller, Yosi; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
In this work we present an approach for matching three-dimensional mesh objects related by isometric transfor- mations and scaling. We propose to utilize the Scale invariant Scale-DoG detector and Local Depth SIFT mesh descriptor, to derive a statistical voting-based scheme to robustly estimate the scale ratio between the registered meshes. This paves the way to formulating a novel non-rigid mesh registration scheme, by matching sets of sparse salient feature points using spectral graph matching. The resulting approach is shown to compare favorably with previous state-of-the-art approaches in registering meshes related by partial alignment, while being a few orders of magnitude faster.
SHREC'12 Track: 3D Mesh Segmentation
(The Eurographics Association, 2012) Lavoué, G.; Vandeborre, J-P.; Benhabiles, H.; Daoudi, M.; Huebner, K.; Mortara, M.; Spagnuolo, M.; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
3D mesh segmentation is a fundamental process in many applications such as shape retrieval, compression, deformation, etc. The objective of this track is to evaluate the performance of recent segmentation methods using a ground-truth corpus and an accurate similarity metric. The ground-truth corpus is composed of 28 watertight models, grouped in five classes (animal, furniture, hand, human and bust) and each associated with 4 ground-truth segmentations done by human subjects. 3 research groups have participated to this track, the accuracy of their segmentation algorithms have been evaluated and compared with 4 other state-of-the-art methods.
Kinematic Skeleton Extraction Based on Motion Boundaries for 3D Dynamic Meshes
(The Eurographics Association, 2012) Benhabiles, Halim; Lavoué, Guillaume; Vandeborre, Jean-Philippe; Daoudi, Mohamed; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
This paper presents a precise kinematic skeleton extraction method for 3D dynamic meshes. Contrary to previous methods, our method is based on the computation of motion boundaries instead of detecting object parts characterized by rigid transformations. Thanks to a learned boundary edge function, we are able to compute efficiently a set of motion boundaries which in fact correspond to all possible articulations of the 3D object. Moreover, the boundaries are detected even if the parts linked to an object's articulation are immobile over time. The different boundaries are then used to extract the kinematic skeleton. Experiments show that our algorithm produces more precise skeletons compared to previous methods.
Effective Shapes Generation for Bayesian CAD Model Reconstruction
(The Eurographics Association, 2012) Bey, Aurelien; Chaine, Raphaelle; Marc, Raphael; Thibault, Guillaume; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
We address the reconstruction of CAD models from 3D point clouds, assuming that an a priori CAD model, roughly similar to the scene to reconstruct, is given. The problem can be solved using a Bayesian approach which states the reconstruction task as the search of the most probable CAD model. This article presents a reliable cylinder generation algorithm that can be used to efficiently reach the configuration which maximizes the target probability, combined with an efficient specific optimization heuristic. Results illustrate how our approach can automatically reconstruct CAD models of industrials scenes.
SHREC'12 Track: Generic 3D Shape Retrieval
(The Eurographics Association, 2012) Li, B.; Godil, A.; Aono, M.; Bai, X.; Furuya, T.; Li, L.; López-Sastre, R.; Johan, H.; Ohbuchi, R.; Redondo-Cabrera, C.; Tatsuma, A.; Yanagimachi, T.; Zhang, S.; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Generic 3D shape retrieval is a fundamental research area in the field of content-based 3D model retrieval. The aim of this track is to measure and compare the performance of generic 3D shape retrieval methods implemented by different participants over the world. The track is based on a new generic 3D shape benchmark, which contains 1200 triangle meshes that are equally classified into 60 categories. In this track, 16 runs have been submitted by 5 groups and their retrieval accuracies were evaluated using 7 commonly used performance metrics.
Parallelized Algorithms for Rigid Surface Alignment on GPU
(The Eurographics Association, 2012) Zabatani, Aviad; Bronstein, Alex M.; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
Alignment and registration of rigid surfaces is a fundamental computational geometric problem with applications ranging from medical imaging, automated target recognition, and robot navigation just to mention a few. The family of the iterative closest point (ICP) algorithms introduced by Chen and Medioni [YC] and Besl and McKey [PB92] and improved over the three decades that followed constitute a classical to the problem. However, with the advent of geometry acquisition technologies and applications they enable, it has become necessary to align in real time dense surfaces containing millions of points. The classical ICP algorithms, being essentially sequential procedures, are unable to address the need. In this study, we follow the recent work by Mitra et al. [NJM] considering ICP from the point of view of point-to-surface Euclidean distance map approximation. We propose a variant of a k-d tree data structure to store the approximation, and show its efficient parallelization on modern graphics processors. The flexibility of our implementation allows using different distance approximation schemes with controllable trade-off between accuracy and complexity. It also allows almost straightforward adaptation to richer transformation groups. Experimental evaluation of the proposed approaches on a state-of-the-art GPU on very large datasets containing around 106 vertices shows real-time performance superior by up to three orders of magnitude compared to an efficient CPU-based version.
Local Goemetrical Feature with Spatial Context for Shape-based 3D Model Retrieval
(The Eurographics Association, 2012) Kawamura, Shun; Usui, Kazuya; Furuya, Takahiko; Ohbuchi, Ryutarou; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
With recent popularity of 3D models, retrieval and recognition of 3D models based on their shape has become an important subject of study. This paper proposes a 3D model retrieval algorithm that is invariant to global deformation as well as to similarity transformation of 3D models. The algorithm is based on a set of local 3D geometrical features combined with bag-of-features approach. The algorithm employs a novel local feature, which is a combination of local geometrical feature enhanced with its spatial context computed as histogram of diffusion distance computed over mesh surface. Experimental evaluation of retrieval accuracy by using benchmark databases showed that adding positional context significantly improves retrieval accuracy.
Towards an Immersive Interface for 3D Object Retrieval
(The Eurographics Association, 2012) Pascoal, Pedro; Ferreira, Alfredo; Jorge, Joaquim; M. Spagnuolo and M. Bronstein and A. Bronstein and A. Ferreira
In this paper we propose the use of immersive virtual reality to visualize and explore query results in 3D object retrieval. These results are shown in a three-dimensional virtual space as 3D objects, organized according to their similarity to the query. Therefore, instead of browsing through a list of thumbnails, the user can explore the query results by navigating among 3D objects in a virtual space and even manipulate these objects for closer a inspection. We performed a preliminary user study and concluded that the proposed solution provides better performance than traditional approaches.

Browse

Browsing 3DOR 12 by Issue Date

Results Per Page

Sort Options