A New 3D-Matching Method of Nonrigid and Partially Similar Models Using Curve Analysis

Tabia, Hedi; Daoudi, Mohamed; Vandeborre, Jean-Philippe; Colot, Olivier

doi:10.1109/tpami.2010.202

Cited by 62 publications

(36 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The precision is kept over 70% when half of the relevant objects have been returned (recall equals to 0.5). Since low recall values correspond to the first objects retrieved, this result shows that our method is significantly [20] 0.918 0.590 0.734 0.841 Tabia et al [34] 0.853 0.527 0.639 0.719 Table 3. Results on SHREC07 for global 3D shape retrieval.…”

Section: Global Shape Retrievalmentioning

confidence: 78%

“…Figure 2 summarizes and compares the precision-recall performance of our approach against three state of the art methods: the Hybrid BoW of Lavoué [20], the curve based method of Tabia et al [34] and the BoW method of Toldo et al [37]. We can clearly see that covariance-based BoW achieves significantly better precision than previous methods for low recall values.…”

Section: Global Shape Retrievalmentioning

confidence: 90%

“…We conducted a quantitative evaluation using the Normalized Discounted Cumulated Gain vector (NDCG) [23]. Figure 4 shows the NDCG curves of the proposed method and six other state-of-the-art methods: the Hybrid BoW [20], the curve-base method [34], the BoW method of Toldo et al [37], the graph-based technique of Tierny et al [36], the extended Reeb graphs (ERG) [3] and the curveskeleton based many-to-many matching (CORNEA) [7]. We can clearly see that our method outperforms them.…”

Section: Partial Shape Retrievalmentioning

confidence: 99%

See 2 more Smart Citations

Covariance Descriptors for 3D Shape Matching and Retrieval

Tabia

Laga

Picard

et al. 2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

100

View full text Add to dashboard Cite

Several descriptors have been proposed in the past for 3D shape analysis, yet none of them achieves best performance on all shape classes. In this paper we propose a novel method for 3D shape analysis using the covariance matrices of the descriptors rather than the descriptors themselves. Covariance matrices enable efficient fusion of different types of features and modalities. They capture, using the same representation, not only the geometric and the spatial properties of a shape region but also the correlation of these properties within the region. Covariance matrices, however, lie on the manifold of Symmetric Positive Definite (SPD) tensors, a special type of Riemannian manifolds, which makes comparison and clustering of such matrices challenging. In this paper we study covariance matrices in their native space and make use of geodesic distances on the manifold as a dissimilarity measure. We demonstrate the performance of this metric on 3D face matching and recognition tasks. We then generalize the Bag of Features paradigm, originally designed in Euclidean spaces, to the Riemannian manifold of SPD matrices. We propose a new clustering procedure that takes into account the geometry of the Riemannian manifold. We evaluate the performance of the proposed Bag of Covariance Matrices framework on 3D shape matching and retrieval applications and demonstrate its superiority compared to descriptor-based techniques.

show abstract

Section: Global Shape Retrievalmentioning

confidence: 78%

Section: Global Shape Retrievalmentioning

confidence: 90%

Section: Partial Shape Retrievalmentioning

confidence: 99%

See 1 more Smart Citation

Covariance Descriptors for 3D Shape Matching and Retrieval

Tabia

Laga

Picard

et al. 2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

100

View full text Add to dashboard Cite

show abstract

“…More exact/interesting algorithms have been applied in 2D e.g. [3], [6], [31], [4] and some in 3D ( [10], [5], [27], [12], [30]). …”

Section: Dense Correspondence Estimationmentioning

confidence: 99%

Class-specific 3D localization using constellations of object parts

Prasad

Knopp

Gool

2011

Procedings of the British Machine Vision Conference 2011

View full text Add to dashboard Cite

Improvement in acquisition systems, has resulted in the ability to capture more realistic 3D models of real world objects, creating a need for better data processing techniques, as in the case of text and images. In this paper, we address the issue of learning classspecific, deformable, 3D part-based structure for object part localization in 3D models/scenes. We employ an inference framework upon fully connected part-based graphs inspired by Pictorial Structures (PS), which combine the local appearance of parts and the long-range structural properties. Using efficient tools for learning the model and performing inference, we show good results on a variety of classes, outperforming PS [7] and ISM [15]. Further, a similar inference framework is employed to find dense correspondences between 3D models, seeded by the above object part localization. Our results show promise for application in more complex 3D processing tasks such as part retrieval, pose estimation, scene understanding and recognition.

show abstract

“…Many problems that appear in these applications can be formulated as the problem of matching and comparing 2D or 3D elastic curves. This includes 3D shape retrieval [14,15], 3D face recognition [10], facial expression recognition [7] and gesture and action recognition [1]. They involve the representation of shapes with curves, the description of the curves with some local or global descriptors and then matching these descriptors using some similarity metric.…”

Section: Introductionmentioning

confidence: 99%

Fast Approximation of Distance Between Elastic Curves using Kernels

Tabia¹,

Picard²,

Laga³

et al. 2013

Procedings of the British Machine Vision Conference 2013

View full text Add to dashboard Cite

Elastic shape analysis on non-linear Riemannian manifolds provides an efficient and elegant way for simultaneous comparison and registration of non-rigid shapes. In such formulation, shapes become points on some high dimensional shape space. A geodesic between two points corresponds to the optimal deformation needed to register one shape onto another. The length of the geodesic provides a proper metric for shape comparison. However, the computation of geodesics, and therefore the metric, is computationally very expensive as it involves a search over the space of all possible rotations and reparameterizations. This problem is even more important in shape retrieval scenarios where the query shape is compared to every element in the collection to search. In this paper, we propose a new procedure for metric approximation using the framework of kernel functions. We will demonstrate that this provides a fast approximation of the metric while preserving its invariance properties.

show abstract

A New 3D-Matching Method of Nonrigid and Partially Similar Models Using Curve Analysis

Cited by 62 publications

References 25 publications

Covariance Descriptors for 3D Shape Matching and Retrieval

Covariance Descriptors for 3D Shape Matching and Retrieval

Class-specific 3D localization using constellations of object parts

Fast Approximation of Distance Between Elastic Curves using Kernels

Contact Info

Product

Resources

About