paper-with-me

홈 › Papers

Boosting Convolutional Neural Networks with Middle Spectrum Grouped Convolution

2023-04-13 · Zhuo Su, Jiehua Zhang, Tianpeng Liu, Zhen Liu, Shuanghui Zhang, Matti Pietikäinen, Li Liu

This paper proposes a novel module called middle spectrum grouped convolution (MSGC) for efficient deep convolutional neural networks (DCNNs) with the mechanism of grouped convolution. It explores the broad "middle spectrum" area between channel pruning and conventional grouped convolution. Compared with channel pruning, MSGC can retain most of the information from the input feature maps due to the group mechanism; compared with grouped convolution, MSGC benefits from the learnability, the core of channel pruning, for constructing its group topology, leading to better channel division. The middle spectrum area is unfolded along four dimensions: group-wise, layer-wise, sample-wise, and attention-wise, making it possible to reveal more powerful and interpretable structures. As a result, the proposed module acts as a booster that can reduce the computational cost of the host backbones for general image recognition with even improved predictive accuracy. For example, in the experiments on ImageNet dataset for image classification, MSGC can reduce the multiply-accumulates (MACs) of ResNet-18 and ResNet-50 by half but still increase the Top-1 accuracy by more than 1%. With 35% reduction of MACs, MSGC can also increase the Top-1 accuracy of the MobileNetV2 backbone. Results on MS COCO dataset for object detection show similar observations. Our code and trained models are available at https://github.com/hellozhuo/msgc.

📄 PDF Abstract BibTeX arXiv:2304.06305

Code (1)

hellozhuo/msgc 공식 구현 pytorch

Tasks

image-classificationImage Classificationobject-detectionObject Detection

Methods 이 논문이 사용한 방법론

Pruning 설명 없음
Pointwise Convolution Pointwise Convolution is a type of convolution that uses a 1x1 kernel: a kernel that iterates through every single point. This…
Batch Normalization 설명 없음
Depthwise Convolution Depthwise Convolution is a type of convolution where we apply a single convolutional filter for each input channel. In the regular 2D…
Depthwise Separable Convolution While standard convolution performs the channelwise and spatial-wise computation in one step, Depthwise Separable Convolution …
Inverted Residual Block 설명 없음
1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…
Average Pooling 설명 없음

Similar Papers 제목 키워드 기반

An Enhanced Scheme for Reducing the Complexity of Pointwise Convolutions in CNNs for Image Classification Based on Interleaved Grouped Filters without Divisibility Constraints

2022-09-08 · Entropy 2022 9 · Joao Paulo Schwarz Schuler, Santiago Romani, Mohamed Abdel-Nasser, Hatem Rashwan 외

In image classification with Deep Convolutional Neural Networks (DCNNs), the number of parameters in pointwise convolutions rapidly grows due to the multiplication of the number of filters by the number of input channels…

image-classificationImage Classification

Beyond Low-pass Filtering: Graph Convolutional Networks with Automatic Filtering

2021-07-10 · Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang 외

Graph convolutional networks are becoming indispensable for deep learning from graph-structured data. Most of the existing graph convolutional networks share two big shortcomings. First, they are essentially low-pass fil…

IoT Malware Detection Architecture using a Novel Channel Boosted and Squeezed CNN

2022-02-08 · Muhammad Asam, Saddam Hussain Khan, Tauseef Jamal, Asifullah Khan

Interaction between devices, people, and the Internet has given birth to a new digital communication model, the Internet of Things (IoT). The seamless network of these smart devices is the core of this IoT model. However…

Malware Detection

Robust Boosting Forests with Richer Deep Feature Hierarchy

2022-10-29 · Jianqiao Wangni

We propose a robust variant of boosting forest to the various adversarial defense methods, and apply it to enhance the robustness of the deep neural network. We retain the deep network architecture, weights, and middle l…

Adversarial DefenseFace Model

GroSS: Group-Size Series Decomposition for Grouped Architecture Search

2019-12-02 · ECCV 2020 8 · Henry Howard-Jenkins, Yiwen Li, Victor A. Prisacariu

We present a novel approach which is able to explore the configuration of grouped convolutions within neural networks. Group-size Series (GroSS) decomposition is a mathematical formulation of tensor factorisation into a …