Representation Learning on Unit Ball with 3D Roto-translational Equivariance

Ramasinghe, Sameera; Khan, Salman; Barnes, Nick; Gould, Stephen J.

doi:10.1007/s11263-019-01278-x

Cited by 9 publications

(7 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The graph is then represented by the Laplacian equation, which is solved using spherical CNNs. Ramasinghe et al [36] investigated the use of radial component in spherical convolutions instead of using spherical convolutions on the sphere surface. They proposed a volumetric convolution operation that was derived from Zernike polynomials.…”

Section: B Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Robust Object Classification Approach Using Spherical Harmonics

et al. 2022

View full text Add to dashboard Cite

Point clouds produced by either 3D scanners or multi-view images are often imperfect and contain noise or outliers. This paper presents an end-to-end robust spherical harmonics approach to classifying 3D objects. The proposed framework first uses the voxel grid of concentric spheres to learn features over the unit ball. We then limit the spherical harmonics order level to suppress the effect of noise and outliers. In addition, the entire classification operation is performed in the Fourier domain. As a result, our proposed model learned features that are less sensitive to data perturbations and corruptions. We tested our proposed model against several types of data perturbations and corruptions, such as noise and outliers. Our results show that the proposed model has fewer parameters, competes with state-of-art networks in terms of robustness to data inaccuracies, and is faster than other robust methods. Our implementation code is also publicly available 1 .INDEX TERMS object recognition, point cloud classification, spherical harmonics, robust classification.

show abstract

Section: B Related Workmentioning

confidence: 99%

“…Compared to 3D CNNs such as the octnet [43], spherical CNNs are rotation equavarient, which could help in increasing our performance. In addition, the use of spherical convolution has shown to have less trainable parameters, where one layer is enough to achieve good performance [36].…”

Section: Implementation Of Spherical Convolutionmentioning

confidence: 99%

Robust Object Classification Approach Using Spherical Harmonics

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The graph is then represented by the Laplacian equation, which is solved using spherical CNNs. Ramasinghe et al [23] investigated the use of radial component in spherical convolutions instead of using spherical convolutions on the sphere surface. They proposed a volumetric convolution operation that was derived from Zernike polynomials.…”

Section: Related Workmentioning

confidence: 99%

“…Compared to 3D CNNs such as the octnet [16], spherical CNNs are rotation equavarient, which could help in increasing our performance. In addition, the use of spherical convolution have shown to have less trainable parameters, where one layer is enough to achieve good performance [23].…”

Section: Implementation Of Spherical Convolutionmentioning

confidence: 99%

Robust Object Classification Approach using Spherical Harmonics

Mukhaimar¹,

Tennakoon²,

Lai³

et al. 2020

Preprint

View full text Add to dashboard Cite

In this paper, we present a robust spherical harmonics approach for the classification of point cloud-based objects. Spherical harmonics have been used for classification over the years, with several frameworks existing in the literature. These approaches use variety of spherical harmonics based descriptors to classify objects. We first investigated these frameworks robustness against data augmentation, such as outliers and noise, as it has not been studied before. Then we propose a spherical convolution neural network framework for robust object classification. The proposed framework uses the voxel grid of concentric spheres to learn features over the unit ball. Our proposed model learn features that are less sensitive to data augmentation due to the selected sampling strategy and the designed convolution operation. We tested our proposed model against several types of data augmentation, such as noise and outliers. Our results show that the proposed model outperforms the state of art networks in terms of robustness to data augmentation.

show abstract

“…T HE use of deep learning for several 3D task such as point cloud classification [1], [2], [3], [4], [5], [6], retrieval [1], [6], and segmentation [2], [7], [8] has shown great success in recent years. However, the success has largely been confined to 3D CAD based benchmarks such as Mod-elNet [1], McGill [9], and Shapenet [10] with very clean data.…”

Section: Introductionmentioning

confidence: 99%

Robust Pooling through the Data Mode

Mukhaimar¹,

Tennakoon²,

Lai³

et al. 2021

Preprint

View full text Add to dashboard Cite

The task of learning from point cloud data is always challenging due to the often occurrence of noise and outliers in the data. Such data inaccuracies can significantly influence the performance of state of the art deep learning networks and their ability to classify or segment objects. While there are some robust deep learning approaches, they are computationally too expensive for real-time applications. This paper proposes a deep learning solution that includes a novel robust pooling layer which greatly enhances network robustness and perform significantly faster than state-of-the-art approaches. The proposed pooling layer looks for data a mode/cluster using two methods, RANSAC, and histogram, as clusters are indicative of models. We tested the pooling layer into frameworks such as Point-based and graph-based neural networks, and the tests showed enhanced robustness as compared to robust state-of-the-art methods.

show abstract

Representation Learning on Unit Ball with 3D Roto-translational Equivariance

Cited by 9 publications

References 56 publications

Robust Object Classification Approach Using Spherical Harmonics

Robust Object Classification Approach Using Spherical Harmonics

Robust Object Classification Approach using Spherical Harmonics

Robust Pooling through the Data Mode

Contact Info

Product

Resources

About