Factorized Convolutional Neural Networks

By Min Wang, Baoyuan Liu, Hassan Foroosh. 2016

188 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. Quoc V. Le
2. Luc Van Gool
3. Dmitry P. Vetrov
4. Hartwig Adam
5. Sergey Levine
6. Weiyang Liu
7. Dacheng Tao
8. Andrew G. Howard

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Dacheng Tao
3. Ming-Hsuan Yang
4. Xiaogang Wang
5. Bernt Schiele
6. Alan L. Yuille
7. Trevor Darrell
8. Andrew Zisserman

In Model Architectures

1. Hartwig Adam
2. Andrew G. Howard
3. Luc Van Gool
4. Menglong Zhu
5. Amos J. Storkey
6. Elliot J. Crowley
7. Gavia Gray
8. Oren Freifeld

Citations in the corpus

• Xception: Deep Learning with Depthwise Separable Convolutions (2016, background): “in 2016 [24] also did related work aiming at reducing the size and computational cost of convolutional neural networks using separable convolutions.”
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications (2017, background): “There has been rising interest in building small and effi-cient neural networks in the recent literature, e.g. [16, 34, 12, 36, 22].”
• Moonshine: Distilling with Cheap Convolutions (2017, background): “Wang et al. (2016) start with separable convolutions and add topological subdivisioning, a way to treat sections of tensors separately, and a bottleneck of the spatial dimensions.”
• In Defense of Pre-Trained ImageNet Architectures for Real-Time Semantic Segmentation of Road-Driving Images (2019, background): “Depthwise separable convolutions [33, 36] decrease computational complexity by splitting a regular convolution in two.”
• Ultimate tensorization: compressing convolutional and FC layers alike (2016, methodology): “This leads to a number of works focusing on compressing and speeding up the convolutional layers [5, 15, 16, 17, 18, 19].”
• Orthogonal Over-Parameterized Training (2020, methodology): “[30, 48, 73] employ sparse and low-rank structures to construct convolution kernels for a efficient neural network.”
• Wavelet Convolutions for Large Receptive Fields (2024, background): “One major change to the traditional convolutions was the introduction of separable convolutions, which separate the dense convolutions into spatial and cross-channel [56, 60].”
• Learning Filter Basis for Convolutional Neural Network Compression (2019, methodology): “Although the compression ratio 1/9 could be achieved by [46], the classification accuracy is severely diminished.”