Open-source alternatives to Efficient Attention

PyTorch implementation of the paper "FcaNet: Frequency Channel Attention Networks".

The official PyTorch code implementation for TIP20' Submission: "Person Re-identification via Attention Pyramid"

This is a PyTorch implementation of CoAtNet specified in "CoAtNet: Marrying Convolution and Attention for All Data Sizes", arXiv 2021.

The PyTorch implementation of Att-DARTS: Differentiable Neural Architecture Search for Attention.

BiSeNetV2 is faster and requires less memory, you can try BiSeNetV2 on cityscapes like this: ` $ export CUDAVISIBLEDEVICES=0,1 $ python -m torch.distributed.launch --nprocpernode=2 bisenetv2/train.py --fp16 ` This would train the model and then compute the mIOU on eval set.

This project is built from IDN, and thanks for the contributions of all the other researchers those who made their codes accessible. paper_link:https://arxiv.org/ftp/arxiv/papers/2104/2104.07566.pdf

This is an official implementation of:

Distilling Knowledge via Knowledge Review

MXNet implementation for:

By Zhongzhan Huang, Senwei Liang, Mingfu Liang and Haizhao Yang.

By Senwei Liang, Zhongzhan Huang (* contribute equally), Mingfu Liang and Haizhao Yang.

Please ★star this repo and cite the following arXiv paper if you think our RNL is useful for you:

This repository provides the code for "CA-Net: Comprehensive attention Convolutional Neural Networks for Explainable Medical Image Segmentation". Our work now is available on Arxivpaperlink. Our work is accepted by TMItmilink.

The Pytorch implementation of "Compact Global Descriptor for Neural Networks" (CGD). arXiv

This is the implementation for PyTroch 0.4.1. - The HRNet OCR version ia available here. - The PyTroch 1.1 version is available here.

by Hengshuang Zhao, Jiaya Jia, and Vladlen Koltun, details are in paper.

By Jie Hu 1 , Li Shen 2 , Samuel Albanie 2 , Gang Sun 1 , Andrea Vedaldi 2 .

Squeeze-and-Excitation Networks

The official PyTorch implementation of "SRM : A Style-based Recalibration Module for Convolutional Neural Networks" for ImageNet. SRM is a lightweight architectural unit that dynamically recalibrates feature responses based on style importance.

PyTorch implementation of Pay Attention to MLPs.

Figure 1. Illustration of our DCANet. We visualize intermediate feature activation using Grad-CAM. Vanilla SE-ResNet50 varies its focus dramatically at different stages. In contrast, our DCA enhanced SE-ResNet50 progressively and recursively adjusts focus, and closely pays attention to the…