Rotation Equivariant Graph Convolutional Network for Spherical Image Classification

Yang, Qin; Li, Chenglin; Dai, Wenrui; Qi, Guo-Jun; Xiong, Hongkai

doi:10.1109/cvpr42600.2020.00436

Cited by 30 publications

(23 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is the way in which the convolution, correction, and pooling layers are interconnected in the processing blocks, as well as the processing blocks between them, which determines the particularity of the network architecture. This architecture is defined as a result of applied research work [42,43].…”

Section: Theoretical Backgroundmentioning

confidence: 99%

Impact of Image Compression on the Performance of Steel Surface Defect Classification with a CNN

Benbarrad

Eloutouate

Arioua

et al. 2021

JSAN

View full text Add to dashboard Cite

Machine vision is increasingly replacing manual steel surface inspection. The automatic inspection of steel surface defects makes it possible to ensure the quality of products in the steel industry with high accuracy. However, the optimization of inspection time presents a great challenge for the integration of machine vision in high-speed production lines. In this context, compressing the collected images before transmission is essential to save bandwidth and energy, and improve the latency of vision applications. The aim of this paper was to study the impact of quality degradation resulting from image compression on the classification performance of steel surface defects with a CNN. Image compression was applied to the Northeastern University (NEU) surface-defect database with various compression ratios. Three different models were trained and tested with these images to classify surface defects using three different approaches. The obtained results showed that trained and tested models on the same compression qualities maintained approximately the same classification performance for all used compression grades. In addition, the findings clearly indicated that the classification efficiency was affected when the training and test datasets were compressed using different parameters. This impact was more obvious when there was a large difference between these compression parameters, and for models that achieved very high accuracy. Finally, it was found that compression-based data augmentation significantly increased the classification precision to perfect scores (98–100%), and thus improved the generalization of models when tested on different compression qualities. The importance of this work lies in exploiting the obtained results to successfully integrate image compression into machine vision systems, and as appropriately as possible.

show abstract

Section: Theoretical Backgroundmentioning

confidence: 99%

Impact of Image Compression on the Performance of Steel Surface Defect Classification with a CNN

Benbarrad

Eloutouate

Arioua

et al. 2021

JSAN

View full text Add to dashboard Cite

show abstract

“…Wang et al [19] then adopt the cubemap representation and unsupervisedly learn monocular 360 • depth estimation. To capture distortion-aware context, several approaches of spherical CNNs are proposed [17,16,4,23,5,7]. With a supervised scheme, Zioulis et al [28] incorporate [17] and propose two network variants to estimate monocular 360 • depth.…”

Section: Related Workmentioning

confidence: 99%

LED2-Net: Monocular 360 Layout Estimation via Differentiable Depth Rendering

Wang,

Yeh,

Sun

et al. 2021

Preprint

View full text Add to dashboard Cite

Although significant progress has been made in room layout estimation, most methods aim to reduce the loss in the 2D pixel coordinate rather than exploiting the room structure in the 3D space. Towards reconstructing the room layout in 3D, we formulate the task of 360 • layout estimation as a problem of predicting depth on the horizon line of a panorama. Specifically, we propose the Differentiable Depth Rendering procedure to make the conversion from layout to depth prediction differentiable, thus making our proposed model end-to-end trainable while leveraging the 3D geometric information, without the need of providing the ground truth depth. Our method achieves state-of-the-art performance on numerous 360 • layout benchmark datasets. Moreover, our formulation enables a pre-training step on the depth dataset, which further improves the generalizability of our layout estimation model.

show abstract

“…For example, Khasanova and Frossard [26] propose a graphbased convolution for ERP in which the edge weights between neighboring nodes (pixels) are equal to the inverse of their distances. Alternatively, other works use a quasi-uniform sampling of the sphere and define graph-based CNNs directly on the sphere [8], [27]. Perraudin et al [8] use Hierarchical Equal Area isoLatitude Pixelation (HEALPix) [28], and Yang et al [27] use Geodesic ICOsahedral Pixelation (GICOPix).…”

Section: Current Solutionsmentioning

confidence: 99%

“…Alternatively, other works use a quasi-uniform sampling of the sphere and define graph-based CNNs directly on the sphere [8], [27]. Perraudin et al [8] use Hierarchical Equal Area isoLatitude Pixelation (HEALPix) [28], and Yang et al [27] use Geodesic ICOsahedral Pixelation (GICOPix). To keep a linear complexity filtering, the convolution is approximated using a Chebyshev polynomial formulation, thus avoiding the graph spectral transform computation.…”

Section: Current Solutionsmentioning

confidence: 99%

OSLO: On-the-Sphere Learning for Omnidirectional images and its application to 360-degree image compression

Bidgoli¹,

Azevedo²,

Maugey³

et al. 2021

Preprint

View full text Add to dashboard Cite

State-of-the-art 2D image compression schemes rely on the power of convolutional neural networks (CNNs). Although CNNs offer promising perspectives for 2D image compression, extending such models to omnidirectional images is not straightforward. First, omnidirectional images have specific spatial and statistical properties that can not be fully captured by current CNN models. Second, basic mathematical operations composing a CNN architecture, e.g., translation and sampling, are not welldefined on the sphere. In this paper, we study the learning of representation models for omnidirectional images and propose to use the properties of HEALPix uniform sampling of the sphere to redefine the mathematical tools used in deep learning models for omnidirectional images. In particular, we: i) propose the definition of a new convolution operation on the sphere that keeps the high expressiveness and the low complexity of a classical 2D convolution; ii) adapt standard CNN techniques such as stride, iterative aggregation, and pixel shuffling to the spherical domain; and then iii) apply our new framework to the task of omnidirectional image compression. Our experiments show that our proposed on-the-sphere solution leads to a better compression gain that can save 13.7% of the bit rate compared to similar learned models applied to equirectangular images. Also, compared to learning models based on graph convolutional networks, our solution supports more expressive filters that can preserve high frequencies and provide a better perceptual quality of the compressed images. Such results demonstrate the efficiency of the proposed framework, which opens new research venues for other omnidirectional vision tasks to be effectively implemented on the sphere manifold.

show abstract

Rotation Equivariant Graph Convolutional Network for Spherical Image Classification

Cited by 30 publications

References 18 publications

Impact of Image Compression on the Performance of Steel Surface Defect Classification with a CNN

Impact of Image Compression on the Performance of Steel Surface Defect Classification with a CNN

LED2-Net: Monocular 360 Layout Estimation via Differentiable Depth Rendering

OSLO: On-the-Sphere Learning for Omnidirectional images and its application to 360-degree image compression

Contact Info

Product

Resources

About