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

Related tags

Deep Learning attention-module
Overview

BAM and CBAM

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

Updates & Notices

  • 2018-10-08: Currently, only CBAM test code is validated. There may be minor errors in the training code. Will be fixed in a few days.
  • 2018-10-11: Training code validated. RESNET50+BAM pretrained weight added.

Requirement

The code is validated under below environment:

  • Ubuntu 16.04, 4*GTX 1080 Ti, Docker (PyTorch 0.4.1, CUDA 9.0 + CuDNN 7.0, Python 3.6)

How to use

ResNet50 based examples are included. Example scripts are included under ./scripts/ directory. ImageNet data should be included under ./data/ImageNet/ with foler named train and val.

# To train with BAM (ResNet50 backbone)
python train_imagenet.py --ngpu 4 --workers 20 --arch resnet --depth 50 --epochs 100 --batch-size 256 --lr 0.1 --att-type BAM --prefix RESNET50_IMAGENET_BAM ./data/ImageNet
# To train with CBAM (ResNet50 backbone)
python train_imagenet.py --ngpu 4 --workers 20 --arch resnet --depth 50 --epochs 100 --batch-size 256 --lr 0.1 --att-type CBAM --prefix RESNET50_IMAGENET_CBAM ./data/ImageNet

Resume with checkpoints

  • ResNet50+CBAM (trained for 100 epochs) checkpoint is provided in this link. ACC@1=77.622 ACC@5=93.948
  • ResNet50+BAM (trained for 90 epochs) checkpoint is provided in this link. ACC@1=76.860 ACC@5=93.416

For validation, please use the script as follows

python train_imagenet.py --ngpu 4 --workers 20 --arch resnet --depth 50 --att-type CBAM --prefix EVAL --resume $CHECKPOINT_PATH$ --evaluate ./data/ImageNet

Other implementations

Comments

  • "the ReLU activation function is followed by W0" Mentioned in your paper is different from your code?

    According to the following code,the W0 is followed by the ReLU activation function? or I misunderstood,cause i'm new to DeepLearningļ¼Œplz pardon me. self.mlp = nn.Sequential( Flatten(), nn.Linear(gate_channels, gate_channels // reduction_ratio), #W0 ? nn.ReLU(), nn.Linear(gate_channels // reduction_ratio, gate_channels) #W1 ? )

    opened by getr1ch 5
  • The configuration about training the CIFAR-100 dataset

    The configuration about training the CIFAR-100 dataset

    At first, thanks you for your code. I want to train the BAM Model on the CIFAR-100 dataset, but I can't find a series of parameters configuration about using the CIFAR-100 dataset. Can you provide the train_cifar100.py, similar to the "train_imagenet.py" Thanks again!

    opened by SWHL 2
  • Can not run in batch size 1

    Can not run in batch size 1

    I try this module in detection task. When i try to make a inference, it throw an error. I found that it can not run forward in one batch size because of the nn.BatchNorm1d . How to deal with it?

    opened by leezonpen 1
  • the model_resnet.py code can not run on CUDA

    the model_resnet.py code can not run on CUDA

    Thanks for sharing the code in the PyTorch version.

    I try to use the model_resnet.py and notice that it can work on the CPU. But when testing it on GPU, I always get an error "IndexError: Dimension out of range (expected to be in the range of [-2, 1], but got 2)" in the line "F.sigmoid(channel_att_sum).unsqueeze(2).unsqueeze(3).expand_as(x)".

    I would like to ask how to solve this problem. Thanks.

    opened by weicyber 0
  • Has anyone run CBAM/BAM on 3D images?

    Has anyone run CBAM/BAM on 3D images?

    Hi, I am working on a classification of 3D MRI where I want to combine a mask and a raw MRI and I was wondering if someone has implemented those models on 3D images

    opened by gkrisp98 0
  • About the pre-trained model

    About the pre-trained model

    Hello,

    Thank you for your great work. The link to the pre-trained model did not seem to work right now. Can you please provide the trained model again?

    Thank you and best regards.

    opened by viet2411 1
  • Combining two attention branches

    Combining two attention branches

    In BAM paper, element-wise summation achieves the best performance. But element-wise product is used in bam.py. Is there any special reason why you get element-wise product?

    opened by ksyoo-git 1
Owner
Jongchan Park
Research Scientist @Lunit
Jongchan Park
GitHub
Official PyTorch code for Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling (HCFlow, ICCV2021)

Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling (HCFlow, ICCV2021) This repository is the official P

Jingyun Liang 159 Dec 30, 2022
Official PyTorch code for Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling (HCFlow, ICCV2021)

Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling (HCFlow, ICCV2021) This repository is the official P

Jingyun Liang 159 Dec 30, 2022
Official PyTorch code of DeepPanoContext: Panoramic 3D Scene Understanding with Holistic Scene Context Graph and Relation-based Optimization (ICCV 2021 Oral).

DeepPanoContext (DPC) [Project Page (with interactive results)][Paper] DeepPanoContext: Panoramic 3D Scene Understanding with Holistic Scene Context G

Cheng Zhang 66 Nov 16, 2022
Official pytorch code for SSAT: A Symmetric Semantic-Aware Transformer Network for Makeup Transfer and Removal

SSAT: A Symmetric Semantic-Aware Transformer Network for Makeup Transfer and Removal This is the official pytorch code for SSAT: A Symmetric Semantic-

ForeverPupil 57 Dec 13, 2022
Official PyTorch Code of GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Monocular 3D Object Detection (CVPR 2021)

GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Monocular 3D Object Detection GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Mo

Abhinav Kumar 76 Jan 2, 2023
Official PyTorch code of Holistic 3D Scene Understanding from a Single Image with Implicit Representation (CVPR 2021)

Implicit3DUnderstanding (Im3D) [Project Page] Holistic 3D Scene Understanding from a Single Image with Implicit Representation Cheng Zhang, Zhaopeng C

Cheng Zhang 149 Jan 8, 2023
Official PyTorch code for CVPR 2020 paper "Deep Active Learning for Biased Datasets via Fisher Kernel Self-Supervision"

Deep Active Learning for Biased Datasets via Fisher Kernel Self-Supervision https://arxiv.org/abs/2003.00393 Abstract Active learning (AL) aims to min

Denis 29 Nov 21, 2022
The official PyTorch code for 'DER: Dynamically Expandable Representation for Class Incremental Learning' accepted by CVPR2021

DER.ClassIL.Pytorch This repo is the official implementation of DER: Dynamically Expandable Representation for Class Incremental Learning (CVPR 2021)

rhyssiyan 108 Jan 1, 2023
Official PyTorch code for WACV 2022 paper "CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows"

CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows WACV 2022 preprint:https://arxiv.org/abs/2107.1

Denis 156 Dec 28, 2022
Official PyTorch code for Mutual Affine Network for Spatially Variant Kernel Estimation in Blind Image Super-Resolution (MANet, ICCV2021)

Mutual Affine Network for Spatially Variant Kernel Estimation in Blind Image Super-Resolution (MANet, ICCV2021) This repository is the official PyTorc

Jingyun Liang 139 Dec 29, 2022
Official pytorch code for SSC-GAN: Semi-Supervised Single-Stage Controllable GANs for Conditional Fine-Grained Image Generation(ICCV 2021)

SSC-GAN_repo Pytorch implementation for 'Semi-Supervised Single-Stage Controllable GANs for Conditional Fine-Grained Image Generation'.PDF SSC-GAN:Sem

tyty 4 Aug 28, 2022
Official Pytorch Code for the paper TransWeather

TransWeather Official Code for the paper TransWeather, Arxiv Tech Report 2021 Paper | Website About this repo: This repo hosts the implentation code,

Jeya Maria Jose 81 Dec 30, 2022
This is the official Pytorch-version code of FlatGCN (Flattened Graph Convolutional Networks for Recommendation).

FlatGCN This is the official Pytorch-version code of FlatGCN (Flattened Graph Convolutional Networks for Recommendation, submitted to ICASSP2022). Req

Dreamer 2 Aug 9, 2022
Official pytorch code for "APP: Anytime Progressive Pruning"

APP: Anytime Progressive Pruning Diganta Misra1,2,3, Bharat Runwal2,4, Tianlong Chen5, Zhangyang Wang5, Irina Rish1,3 1 Mila - Quebec AI Institute,2 L

Landskape AI 12 Nov 22, 2022
Sound-guided Semantic Image Manipulation - Official Pytorch Code (CVPR 2022)

?? Sound-guided Semantic Image Manipulation (CVPR2022) Official Pytorch Implementation Sound-guided Semantic Image Manipulation IEEE/CVF Conference on

CVLAB 58 Dec 28, 2022
[CVPR 2022] Official Pytorch code for OW-DETR: Open-world Detection Transformer

OW-DETR: Open-world Detection Transformer (CVPR 2022) [Paper] Akshita Gupta*, Sanath Narayan*, K J Joseph, Salman Khan, Fahad Shahbaz Khan, Mubarak Sh

Akshita Gupta 127 Dec 27, 2022
Plug and play transformer you can find network structure and official complete code by clicking List

Plug-and-play Module Plug and play transformer you can find network structure and official complete code by clicking List The following is to quickly

null 8 Mar 27, 2022
Amazon Forest Computer Vision: Satellite Image tagging code using PyTorch / Keras with lots of PyTorch tricks

Amazon Forest Computer Vision Satellite Image tagging code using PyTorch / Keras Here is a sample of images we had to work with Source: https://www.ka

Mamy Ratsimbazafy 360 Dec 10, 2022
Amazon Forest Computer Vision: Satellite Image tagging code using PyTorch / Keras with lots of PyTorch tricks

Amazon Forest Computer Vision Satellite Image tagging code using PyTorch / Keras Here is a sample of images we had to work with Source: https://www.ka

Mamy Ratsimbazafy 359 Jan 5, 2023