Multigrid Neural Architectures

By Tsung-Wei Ke, Michael Maire, Stella X. Yu. 2016

99 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Model Architectures

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jiashi Feng
2. Quoc V. Le
3. Shuicheng Yan
4. Trevor Darrell
5. Alan L. Yuille
6. Dacheng Tao
7. Luc Van Gool
8. Andrea Vedaldi

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Jiashi Feng
3. Dacheng Tao
4. Xiaogang Wang
5. Trevor Darrell
6. Ming-Hsuan Yang
7. Andrea Vedaldi
8. Alan L. Yuille

In Model Architectures

1. Jiashi Feng
2. Shuicheng Yan
3. Quoc V. Le
4. Yunpeng Chen
5. Christoph Feichtenhofer
6. Alan L. Yuille
7. Yannis Kalantidis
8. Gao Huang

Citations in the corpus

• Benchmarking Neural Network Robustness to Common Corruptions and Perturbations (2019, background): “Some multiscale architectures are called Multigrid Networks (Ke et al., 2017).”
• Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks (2018, background): “The Inception family of architectures [38, 37] popularised the use of multi-scale convolutional modules, which help ensure the efficient aggregation of context throughout the hierarchy of learned representations [17].”
• Resolution Adaptive Networks for Efficient Inference (2020, background): “This further agrees with the “coarse to fine processing” efficient algorithm design in [17].”
• Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks With Octave Convolution (2019, methodology): “The multigrid conv (MG-Conv) [25] is a bi-directional and crossscale convolution operator.”
• Multiresolution Tree Networks for 3D Point Cloud Processing (2018, methodology): “[22] showed that communication across multiresolutions of an image throughout the network leads to improved convergence and better accuracy on a variety of tasks.”
• Deciding How to Decide: Dynamic Routing in Artificial Neural Networks (2017, methodology): “To address this, instead of computing a 2-dimensional array of local features at each layer, we compute a pyramid of features (resembling the pyramids described by Ke et al. (2016)), with local descriptors at the bottom and global descriptors at the top.”
• Benchmarking Neural Network Robustness to Common Corruptions and Surface Variations (2018, background): “es across scales at each layer rather than slowly gaining a global representation of the input as in typical convolutional neural networks. Some multiscale architectures are called Multigrid Networks [34]. Multigrid networks each have a pyramid of grids in each layer which enables the subsequent lay”