Attention Branch Network

Writer : Hiroshi Fukui

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. This link is the arXiv paper.

Citation

@article{fukui2018,
	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{fukui2018,  
	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}  
}  

Detail

Our source code is based on https://github.com/bearpaw/pytorch-classification/ with PyTorch. Requirements of PyTorch version is as follows, and we published the Docker file. If you need the Docker, please use our Dockerfile.

• PyTorch : 0.4.0
• PyTorch vision : 0.2.1

Example of run command is as follows, but if you try the other models or ImageNet, please refer the README at https://github.com/bearpaw/pytorch-classification/blob/master/TRAINING.md.

• Training

python 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

python imagenet.py -a resnet152 --data ../../../../dataset/imagenet_data/ --epochs 90 --schedule 31 61 --gamma 0.1 -c checkpoints/imagenet/resnet152 --gpu-id 0,1,2,3 --test-batch 100

• Evaluation

python 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

python 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

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

Models on CIFAR100 Dataset

• ResNet110 : https://www.dropbox.com/sh/6mkk6e9c4qanznz/AACsvn_52Evk9ONjM5yWW8Yra?dl=0
• DenseNet : https://www.dropbox.com/sh/j572oriksv30g4n/AACoH7pQ8sntnZUr0kU2LNbLa?dl=0
• Wide ResNet : https://www.dropbox.com/sh/ou4nbl4mii54p0e/AACwN41B584l9yUzN7CAX_8ja?dl=0
• ResNeXt : https://www.dropbox.com/sh/ds0uk0ileffad7h/AAC-nILHLRzFEy06ohBkVOVha?dl=0

Models on ImageNet2012 Dataset

• ResNet50 : https://www.dropbox.com/sh/tuo90s1uqmbk1vd/AAAksM9uPT5u-eViAe-PXIqsa?dl=0
• ResNet101 : https://www.dropbox.com/sh/8vzv7ov59xb5wle/AABaE24vo3Kc-VuSKbOUsZiua?dl=0
• ResNet152 : https://www.dropbox.com/sh/senw3akoud9cten/AACXOBuiNCWq6wPjl4EIVmhHa?dl=0

Performance of the ABN on 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

Performance of the ABN on ImageNet2012

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

Example of Attention map