CBAM-tensorflow-slim

CBAM-TensorFlow-Slim

This is a Tensorflow implementation of “CBAM: Convolutional Block Attention Module” aiming to be compatible on the TensorFlow-Slim image classification model library. This repository includes the implementation of “SENet-tensorflow-slim”.

If you want to use simpler implementation, check the repository “CBAM-TensorFlow” which includes simple tensorflow implementation of ResNext, Inception-V4, and Inception-ResNet-V2 on Cifar10 dataset.

CBAM: Convolutional Block Attention Module

CBAM proposes an architectural unit called “Convolutional Block Attention Module” (CBAM) block to improve representation power by using attention mechanism: focusing on important features and supressing unnecessary ones. This research can be considered as a descendant and an improvement of “Squeeze-and-Excitation Networks”.

Diagram of a CBAM_block

Diagram of each attention sub-module

Classification results on ImageNet-1K

Prerequisites

• Python 3.x
• TensorFlow 1.x
• TF-slim
  • Check the ‘installation’ part of TF-Slim image models README.

Prepare Data set

You should prepare your own dataset or open dataset (Cifar10, flowers, MNIST, ImageNet). For preparing dataset, you can follow the ‘preparing the datasets’ part in TF-Slim image models README.

CBAM_block and SE_block Supportive Models

This project is based on TensorFlow-Slim image classification model library. Every image classification model in TensorFlow-Slim can be run the same. And, you can run CBAM_block or SE_block added models in the below list by adding one argument --attention_module=cbam_block or --attention_module=se_block when you train or evaluate a model.

• Inception V4 + CBAM / + SE
• Inception-ResNet-v2 + CBAM / + SE
• ResNet V1 50 + CBAM / + SE
• ResNet V1 101 + CBAM / + SE
• ResNet V1 152 + CBAM / + SE
• ResNet V1 200 + CBAM / + SE
• ResNet V2 50 + CBAM / + SE
• ResNet V2 101 + CBAM / + SE
• ResNet V2 152 + CBAM / + SE
• ResNet V2 200 + CBAM / + SE

Change Reduction ratio

To change reduction ratio, you have to manually set the ratio on def cbam_block(input_feature, name, ratio=16) method for cbam_block or def se_block(residual, name, ratio=8) method for se_block in CBAM-tensorflow-slim/nets/attention_module.py.

Train a Model

You can find example of training script in CBAM-tensorflow-slim/scripts/.

Train a model with CBAM_block

Below script gives you an example of training a model with CBAM_block.

DATASET_DIR=/DIRECTORY/TO/DATASET
TRAIN_DIR=/DIRECTORY/TO/TRAIN
CUDA_VISIBLE_DEVICES=0 python train_image_classifier.py \
    --train_dir=${TRAIN_DIR} \
    --dataset_name=imagenet \
    --dataset_split_name=train \
    --dataset_dir=${DATASET_DIR} \
    --model_name=resnet_v1_50 \
    --batch_size=100 \
    --attention_module=cbam_block

Train a model with SE_block

Below script gives you an example of training a model with SE_block.

DATASET_DIR=/DIRECTORY/TO/DATASET
TRAIN_DIR=/DIRECTORY/TO/TRAIN
CUDA_VISIBLE_DEVICES=0 python train_image_classifier.py \
    --train_dir=${TRAIN_DIR} \
    --dataset_name=imagenet \
    --dataset_split_name=train \
    --dataset_dir=${DATASET_DIR} \
    --model_name=resnet_v1_50 \
    --batch_size=100 \
    --attention_module=se_block

Train a model without attention module

Below script gives you an example of training a model without attention module.

DATASET_DIR=/DIRECTORY/TO/DATASET
TRAIN_DIR=/DIRECTORY/TO/TRAIN
CUDA_VISIBLE_DEVICES=0 python train_image_classifier.py \
    --train_dir=${TRAIN_DIR} \
    --dataset_name=imagenet \
    --dataset_split_name=train \
    --dataset_dir=${DATASET_DIR} \
    --model_name=resnet_v1_50 \
    --batch_size=100

Evaluate a Model

You can find example of evaluation script in CBAM-tensorflow-slim/scripts/. To keep track of validation accuracy while training, you can use eval_image_classifier_loop.py which evaluate the performance at multiple checkpoints during training. If you want to just evaluate a model once, you can use eval_image_classifier.py.

Evaluate a model with CBAM_block

Below script gives you an example of evaluating a model with CBAM_block during training.

DATASET_DIR=/DIRECTORY/TO/DATASET
CHECKPOINT_FILE=/DIRECTORY/TO/CHECKPOINT
EVAL_DIR=/DIRECTORY/TO/EVAL
CUDA_VISIBLE_DEVICES=0 python eval_image_classifier_loop.py \
    --alsologtostderr \
    --checkpoint_path=${CHECKPOINT_FILE} \
    --dataset_dir=${DATASET_DIR} \
    --eval_dir=${EVAL_DIR} \
    --dataset_name=imagenet \
    --dataset_split_name=validation \
    --model_name=resnet_v1_50 \
    --batch_size=100 \
    --attention_module=cbam_block

Evaluate a model with SE-block

Below script gives you an example of evaluating a model with SE_block during training.

DATASET_DIR=/DIRECTORY/TO/DATASET
CHECKPOINT_FILE=/DIRECTORY/TO/CHECKPOINT
EVAL_DIR=/DIRECTORY/TO/EVAL
CUDA_VISIBLE_DEVICES=0 python eval_image_classifier_loop.py \
    --alsologtostderr \
    --checkpoint_path=${CHECKPOINT_FILE} \
    --dataset_dir=${DATASET_DIR} \
    --eval_dir=${EVAL_DIR} \
    --dataset_name=imagenet \
    --dataset_split_name=validation \
    --model_name=resnet_v1_50 \
    --batch_size=100 \
    --attention_module=se_block

Evaluate a model without attention module

Below script gives you an example of evaluating a model without attention module during training.

DATASET_DIR=/DIRECTORY/TO/DATASET
CHECKPOINT_FILE=/DIRECTORY/TO/CHECKPOINT
EVAL_DIR=/DIRECTORY/TO/EVAL
CUDA_VISIBLE_DEVICES=0 python eval_image_classifier_loop.py \
    --alsologtostderr \
    --checkpoint_path=${CHECKPOINT_FILE} \
    --dataset_dir=${DATASET_DIR} \
    --eval_dir=${EVAL_DIR} \
    --dataset_name=imagenet \
    --dataset_split_name=validation \
    --model_name=resnet_v1_50 \
    --batch_size=100 

Related Works

• Blog: CBAM: Convolutional Block Attention Module
• Repository: CBAM-TensorFlow
• Repository: CBAM-Keras
• Repository: SENet-TensorFlow-Slim

Reference

• Paper: CBAM: Convolutional Block Attention Module
• Paper: Squeeze-and-Excitation Networks
• Repository: TensorFlow-Slim image classification model library

Author

Byung Soo Ko / [email protected]