Attention Branch Network

Writer : Hiroshi Fukui

Maintainer : Tsubasa Hirakawa

Abstract

This repository contains the source code of Attention Branch Network for image classification. The attention branch network is designed to extend the top-down visual explanation model by introducing a branch structure with an attention mechanism. ABN improves CNN’s performance and visual explanation at the same time by gnerating attention maps in the forward pass.

[CVF open access] [arXiv paper]

Citation

If you find this repository is useful. Please cite the following references.

@article{fukui2018cvpr,
    author = {Hiroshi Fukui and Tsubasa Hirakawa and Takayoshi Yamashita and Hironobu Fujiyoshi},
    title = {Attention Branch Network: Learning of Attention Mechanism for Visual Explanation},
    journal = {Computer Vision and Pattern Recognition},
    year = {2019},
    pages = {10705-10714}
}
@article{fukui2018arxiv,
    author = {Hiroshi Fukui and Tsubasa Hirakawa and Takayoshi Yamashita and Hironobu Fujiyoshi},
    title = {Attention Branch Network: Learning of Attention Mechanism for Visual Explanation},
    journal = {arXiv preprint arXiv:1812.10025},
    year = {2018}
}

Enviroment

Our source code is based on https://github.com/bearpaw/pytorch-classification/ implemented with PyTorch. We are grateful for the author! Requirements of PyTorch version are as follows:

Docker

We prepared Docker environments for ABN. You quickly start to use Docker and run scripts. For more details, please see docker/README.md.

Execution

Example of run command is as follows:

Training

python3 cifar.py -a resnet --dataset cifar100 --depth 110 --epochs 300 --schedule 150 225 --gamma 0.1 --wd 1e-4 --checkpoint checkpoints/cifar100/resnet-110 --gpu-id 0,1

python3 imagenet.py -a resnet152 --data ../../dataset/imagenet_data/ --epochs 90 --schedule 31 61 --gamma 0.1 -c checkpoints/imagenet/resnet152 --gpu-id 4,5,6,7 --test-batch 100

Evaluation

python3 cifar.py -a resnet --dataset cifar100 --depth 110 --epochs 300 --schedule 150 225 --gamma 0.1 --wd 1e-4 --checkpoint checkpoints/cifar100/resnet-110 --gpu-id 0,1 --evaluate --resume checkpoints/cifar100/resnet-110/model_best.pth.tar

python3 imagenet.py -a resnet152 --data ../../../../dataset/imagenet_data/ --epochs 90 --schedule 31 61 --gamma 0.1 -c checkpoints/imagenet/resnet152 --gpu-id 4,5,6 --test-batch 10 --evaluate --resume checkpoints/imagenet/resnet152/model_best.pth.tar

If you try the other models or ImageNet, please see TRAINING.md.

Trained models

We have published the model files of ABN, which are ResNet family models on CIFAR100 and ImageNet2012 dataset.

Models on CIFAR100 Dataset

Models on ImageNet2012 Dataset

Performances

CIFAR100

top-1 error (ABN) top-1 error (original)
ResNet110 22.5 24.1
DenseNet 21.6 22.5
Wide ResNet 18.1 18.9
ResNeXt 17.7 18.3

ImageNet2012

top-1 error (ABN) top-1 error (original)
ResNet50 23.1 24.1
ResNet101 21.8 22.5
ResNet152 21.4 22.2

Examples of attention map

overview image