Pose normalization of 3D models via reflective symmetry on panoramic views

Sfikas, Konstantinos; Theoharis, Theoharis; Pratikakis, Ioannis

doi:10.1007/s00371-014-0935-4

Cited by 13 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The motivation behind the aforementioned method is that the PANORAMA representation is able to bridge the dimensionality gap between 3D object space and the 2D image input that is typically suitable for a convolutional neural network, in a very efficient manner. PANORAMA has already proven to be a successful hand-crafted 3D model descriptor that has achieved state-of-the-art 3D model retrieval performance in various implementations, (Papadakis et al, 2010;Sfikas et al, 2014Sfikas et al, , 2013aSfikas et al, , 2016. It has also been used as input to a successful pose normalization method, SYMPAN (Sfikas et al, 2014) (briefly detailed in 3.1.2).…”

mentioning

confidence: 99%

“…PANORAMA has already proven to be a successful hand-crafted 3D model descriptor that has achieved state-of-the-art 3D model retrieval performance in various implementations, (Papadakis et al, 2010;Sfikas et al, 2014Sfikas et al, , 2013aSfikas et al, , 2016. It has also been used as input to a successful pose normalization method, SYMPAN (Sfikas et al, 2014) (briefly detailed in 3.1.2). This work constitutes an extension of the method presented in (Sfikas et al, 2017).…”

mentioning

confidence: 99%

“…In (Sfikas et al, 2013b(Sfikas et al, , 2014 a pose normalization 64 method, SYMPAN, based on reflective symmetry computed on 65 PANORAMA-based views, is presented. Initially, through an 66 iterative procedure, the symmetry principal plane of a 3D model 67 is estimated, thus computing the first axis of the model.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Ensemble of PANORAMA-based convolutional neural networks for 3D model classification and retrieval

Sfikas

Pratikakis

Theoharis

2018

Computers & Graphics

Self Cite

134

View full text Add to dashboard Cite

A novel method for the classification and retrieval of 3D models is proposed; it exploits the 2D panoramic view representation of 3D models as input to an ensemble of Convolutional Neural Networks which automatically compute the features. The first step of the proposed pipeline, pose normalization is performed using the SYMPAN method, which is also computed on the panoramic view representation. In the training phase, three panoramic views corresponding to the major axes, are used for the training of an ensemble of Convolutional Neural Networks. The panoramic views consist of 3-channel images, containing the Spatial Distribution Map, the Normals' Deviation Map and the magnitude of the Normals' Devation Map Gradient Image. The proposed method aims at capturing feature continuity of 3D models, while simultaneously minimizing data preprocessing via the construction of an augmented image representation. It is extensively tested in terms of classification and retrieval accuracy on two standard large scale datasets: ModelNet and ShapeNet. 1. Introduction 1 In the recent past, convolutional neural networks (CNN) have 2 shown their superiority against humans in computing features, 3 while they are very sensitive to the input representation. In this 4 work an extension of the PANORAMA 3D shape representa-5 tion, previously proposed by our team (Papadakis et al., 2010), 6 is exploited as the input representation to a CNN for computing 7 descriptor features for 3D object classification and retrieval. 8 The 3D models are initially pose normalized using the SYM-9 PAN pose normalization algorithm, (Sfikas et al., 2014) which 10 is based on the use of reflective symmetry on their panoramic 11 view images. Next, an augmented panoramic view is created 12 and used to train the convolutional neural network. This aug-13 mented panoramic view consists of the spatial and orientation 14 components of PANORAMA, (see 3.1.1), along with the mag-15 nitude of the gradient image which is extracted from the ori-16 entation component. A reduction in the size of the augmented 17 panoramic view representation is shown to benefit the training 18 procedure.

show abstract

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Ensemble of PANORAMA-based convolutional neural networks for 3D model classification and retrieval

Sfikas

Pratikakis

Theoharis

2018

Computers & Graphics

Self Cite

134

View full text Add to dashboard Cite

show abstract

“…The computation of the translation is usually done by aligning barycentres, while the scaling is done by aligning bounding boxes or bounding spheres (Sfikas et al, 2014). These are both considered less challenging to compute than the rotation, which is the focus of this paper.…”

Section: Introductionmentioning

confidence: 99%

Mesh Alignment using Grid based PCA

Kaye¹,

Ivrissimtzis²

2015

Proceedings of the 10th International Conference on Computer Graphics Theory and Applications

View full text Add to dashboard Cite

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-pro t purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details. An algorithm is presented to align one mesh to another by means of a regular 3D grid upon which Principal Components Analysis (PCA) is performed. The use of a 3D lattice external to both inputs increases the robustness of PCA, particularly when dealing with meshes of different and possibly uneven vertex density. The proposed algorithm was tested on meshes that have undergone standard mesh processing operations such as smoothing, simplification and remeshing. In several cases the results indicate an improved robustness compared to performing PCA directly on mesh vertices.

show abstract

Object restoration based on extrinsic reflective symmetry plane detection

Cao

Liu

et al. 2021

Vis Comput

View full text Add to dashboard Cite

Pose normalization of 3D models via reflective symmetry on panoramic views

Cited by 13 publications

References 36 publications

Ensemble of PANORAMA-based convolutional neural networks for 3D model classification and retrieval

Ensemble of PANORAMA-based convolutional neural networks for 3D model classification and retrieval

Mesh Alignment using Grid based PCA

Object restoration based on extrinsic reflective symmetry plane detection

Contact Info

Product

Resources

About