What are the main features of jongchan/attention-module?

The main features of jongchan/attention-module are: Attention Mechanisms, Computer Vision, Image Classification Architectures.

What are some open-source alternatives to jongchan/attention-module?

Open-source alternatives to jongchan/attention-module include: hujie-frank/senet — Squeeze-and-Excitation Networks. kaiminghe/resnet-1k-layers — Deep Residual Networks with 1K Layers. facebookresearch/resnext — Implementation of a classification framework from the paper Aggregated Residual Transformations for Deep Neural Networks. liuzhuang13/densenet — DenseNet is a computer vision model and convolutional neural network implementation designed for image recognition and… implus/sknet. pytorch/vision — This project is a comprehensive computer vision library for the PyTorch ecosystem, providing a standardized collection…

Attention Module - add attention blocks

Name: jongchan/attention-module
Author: Jongchan

Official PyTorch code for "BAM: Bottleneck Attention Module (BMVC2018)" and "CBAM: Convolutional Block Attention Module (ECCV2018)"

Features

Attention Mechanisms - Convolutional block and bottleneck attention modules.
Computer Vision - Convolutional block attention module for feature refinement.
Image Classification Architectures - Attention module implementations for convolutional networks.

Official PyTorch code for "BAM: Bottleneck Attention Module (BMVC2018)" and "CBAM: Convolutional Block Attention Module (ECCV2018)"

Features

Attention Mechanisms - Convolutional block and bottleneck attention modules.
Computer Vision - Convolutional block attention module for feature refinement.
Image Classification Architectures - Attention module implementations for convolutional networks.

hujie-frank/SENet

3,641View on GitHub

Squeeze-and-Excitation Networks

implus/SKNet

0View on GitHub

facebookresearch/ResNeXt

1,926View on GitHub

Implementation of a classification framework from the paper Aggregated Residual Transformations for Deep Neural Networks

KaimingHe/resnet-1k-layers

936View on GitHub

Deep Residual Networks with 1K Layers