Spherical Convolution

Spherical convolution is a technique extending convolutional neural networks (CNNs) to process data defined on the surface of a sphere, addressing the limitations of traditional CNNs in handling non-planar data. Current research focuses on improving the efficiency and accuracy of spherical CNNs, particularly within applications like diffusion MRI analysis (using architectures such as U-Nets and incorporating spherical deconvolution) and 6D object pose estimation (employing novel spherical convolution designs and incorporating graph convolutional networks). These advancements are significantly impacting fields like medical imaging (improving brain microstructure analysis) and computer vision (enhancing object recognition and pose estimation), enabling more accurate and efficient processing of spherical data.

Papers

November 18, 2024

Equivariant spatio-hemispherical networks for diffusion MRI deconvolution
Axel Elaldi, Guido Gerig, Neel Dey
High Dimensional Equivariant Network Image Deconvolution White Matter Diffusion MRI Cranial Nerve Tractography Spherical Convolution

September 2, 2024

Ground-truth effects in learning-based fiber orientation distribution estimation in neonatal brains
Rizhong Lin, Hamza Kebiri, Ali Gholipour, Yufei Chen, Jean-Philippe Thiran, Davood Karimi, Meritxell Bach Cuadra
Ground Truth White Matter Voxel Based Semantic Segmentation Infant Brain Spherical Convolution Fiber Orientation Distribution

August 19, 2023

VI-Net: Boosting Category-level 6D Object Pose Estimation via Learning Decoupled Rotations on the Spherical Representations
Jiehong Lin, Zewei Wei, Yabin Zhang, Kui Jia
6D Object Rotation Angle Rotation Estimation W Net Spherical Convolutional Neural Network Category Level 6D Object Pose Spherical Convolution Spherical Representation

July 28, 2023

Recovering high-quality FODs from a reduced number of diffusion-weighted images using a model-driven deep learning architecture
J Bartlett, C E Davey, L A Johnston, J Duan
High Quality Constructive Reduction Channel Acquisition Diffusion Weighted Spherical Convolution Fiber Orientation Distribution Model Driven Deep

June 8, 2023

Scaling Spherical CNNs
Carlos Esteves, Jean-Jacques Slotine, Ameesh Makadia
Spherical Convolutional Neural Network Traditional Convolution Spherical Convolution

June 5, 2023

Robust Fiber ODF Estimation Using Deep Constrained Spherical Deconvolution for Diffusion MRI
Tianyuan Yao, Francois Rheault, Leon Y Cai, Vishwesh nath, Zuhayr Asad, Nancy Newlin, Can Cui, Ruining Deng, Karthik Ramadass, Andrea Shafer, Susan Resnick, Kurt Schilling, Bennett A. Landman, Yuankai Huo
Diffusion MRI Longitudinal Study Spherical Convolution Fiber Orientation Distribution

May 5, 2023

Physics-based network fine-tuning for robust quantitative susceptibility mapping from high-pass filtered phase
Jinwei Zhang, Alexey Dimov, Chao Li, Hang Zhang, Thanh D. Nguyen, Pascal Spincemaille, Yi Wang
Fine Tuning Physic Informed Civil Engineering Phase Spherical Convolution Quantitative Susceptibility Mapping

April 12, 2023

$E(3) \times SO(3)$-Equivariant Networks for Spherical Deconvolution in Diffusion MRI
Axel Elaldi, Guido Gerig, Neel Dey
Equivariant Network Diffusion MRI Spherical Convolution Late Time

September 27, 2022

Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions
Jeremy Ocampo, Matthew A. Price, Jason D. McEwen
Potential Scalability Direct Convolution Continuous Variable Group Convolution Spherical Convolutional Neural Network Spherical Convolution

July 21, 2022

Deep Audio Waveform Prior
Arnon Turetzky, Tzvi Michelson, Yossi Adi, Shmuel Peleg
Audio Source Separation Deep Prior Spherical Convolution Audio Waveform

June 3, 2022

Spatial Feature Mapping for 6DoF Object Pose Estimation
Jianhan Mei, Xudong Jiang, Henghui Ding
Graph Convolutional Network Object Pose Estimation 3D Keypoints Feature Mapping Spherical Convolution 2 Dimensional Feature

January 29, 2022

Spherical Convolution empowered FoV Prediction in 360-degree Video Multicast with Limited FoV Feedback
Jie Li, Ling Han, Cong Zhang, Qiyue Li, Zhi Liu
Field of View Spherical Convolution

December 23, 2021

NeRD++: Improved 3D-mirror symmetry learning from a single image
Yancong Lin, Silvia-Laura Pintea, Jan van Gemert
Single Image Cyclic Symmetry Unseen 3D Pixel Correspondence Spherical Convolution Mirror Detection

Spherical Convolution

Papers

Equivariant spatio-hemispherical networks for diffusion MRI deconvolution

Ground-truth effects in learning-based fiber orientation distribution estimation in neonatal brains

VI-Net: Boosting Category-level 6D Object Pose Estimation via Learning Decoupled Rotations on the Spherical Representations

Recovering high-quality FODs from a reduced number of diffusion-weighted images using a model-driven deep learning architecture

Scaling Spherical CNNs

Robust Fiber ODF Estimation Using Deep Constrained Spherical Deconvolution for Diffusion MRI

Physics-based network fine-tuning for robust quantitative susceptibility mapping from high-pass filtered phase

$E(3) \times SO(3)$-Equivariant Networks for Spherical Deconvolution in Diffusion MRI

Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions

Deep Audio Waveform Prior

Spatial Feature Mapping for 6DoF Object Pose Estimation

Spherical Convolution empowered FoV Prediction in 360-degree Video Multicast with Limited FoV Feedback

NeRD++: Improved 3D-mirror symmetry learning from a single image