RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition

05/05/2021
by   Xiaohan Ding, et al.
0

We propose RepMLP, a multi-layer-perceptron-style neural network building block for image recognition, which is composed of a series of fully-connected (FC) layers. Compared to convolutional layers, FC layers are more efficient, better at modeling the long-range dependencies and positional patterns, but worse at capturing the local structures, hence usually less favored for image recognition. We propose a structural re-parameterization technique that adds local prior into an FC to make it powerful for image recognition. Specifically, we construct convolutional layers inside a RepMLP during training and merge them into the FC for inference. On CIFAR, a simple pure-MLP model shows performance very close to CNN. By inserting RepMLP in traditional CNN, we improve ResNets by 1.8 2.3 that combining the global representational capacity and positional perception of FC with the local prior of convolution can improve the performance of neural network with faster speed on both the tasks with translation invariance (e.g., semantic segmentation) and those with aligned images and positional patterns (e.g., face recognition). The code and models are available at https://github.com/DingXiaoH/RepMLP.

READ FULL TEXT
research
12/21/2021

RepMLPNet: Hierarchical Vision MLP with Re-parameterized Locality

Compared to convolutional layers, fully-connected (FC) layers are better...
research
03/18/2019

Offline and Online Deep Learning for Image Recognition

Image recognition using Deep Learning has been evolved for decades thoug...
research
12/29/2022

BiMLP: Compact Binary Architectures for Vision Multi-Layer Perceptrons

This paper studies the problem of designing compact binary architectures...
research
07/09/2015

Understanding Intra-Class Knowledge Inside CNN

Convolutional Neural Network (CNN) has been successful in image recognit...
research
07/04/2018

Restructuring Batch Normalization to Accelerate CNN Training

Because CNN models are compute-intensive, where billions of operations c...
research
05/17/2021

Itsy Bitsy SpiderNet: Fully Connected Residual Network for Fraud Detection

With the development of high technology, the scope of fraud is increasin...
research
10/09/2013

Neural perceptual model to global-local vision for recognition of the logical structure of administrative documents

This paper gives the definition of Transparent Neural Network "TNN" for ...

Please sign up or login with your details

Forgot password? Click here to reset