The official implementation of ELSA: Enhanced Local Self-Attention for Vision Transformer

DamoCV

Last update: Dec 19, 2022

Related tags

Deep Learning ELSA

Overview

ELSA: Enhanced Local Self-Attention for Vision Transformer

By Jingkai Zhou, Pichao Wang*, Fan Wang, Qiong Liu, Hao Li, Rong Jin

This repo is the official implementation of "ELSA: Enhanced Local Self-Attention for Vision Transformer".

Introduction

Self-attention is powerful in modeling long-range dependencies, but it is weak in local finer-level feature learning. As shown in Figure 1, the performance of local self-attention (LSA) is just on par with convolution and inferior to dynamic filters, which puzzles researchers on whether to use LSA or its counterparts, which one is better, and what makes LSA mediocre. In this work, we comprehensively investigate LSA and its counterparts. We find that the devil lies in the generation and application of spatial attention.

Based on these findings, we propose the enhanced local self-attention (ELSA) with Hadamard attention and the ghost head, as illustrated in Figure 2. Experiments demonstrate the effectiveness of ELSA. Without architecture / hyperparameter modification, The use of ELSA in drop-in replacement boosts baseline methods consistently in both upstream and downstream tasks.

Please refer to our paper for more details.

Model zoo

ImageNet Classification

Model	#Params	Pretrain	Resolution	Top1 Acc	Download
ELSA-Swin-T	28M	ImageNet 1K	224	82.7	google / baidu
ELSA-Swin-S	53M	ImageNet 1K	224	83.5	google / baidu
ELSA-Swin-B	93M	ImageNet 1K	224	84.0	google / baidu

COCO Object Detection

Backbone	Method	Pretrain	Lr Schd	Box mAP	Mask mAP	#Params	Download
ELSA-Swin-T	Mask R-CNN	ImageNet-1K	1x	45.7	41.1	49M	google / baidu
ELSA-Swin-T	Mask R-CNN	ImageNet-1K	3x	47.5	42.7	49M	google / baidu
ELSA-Swin-S	Mask R-CNN	ImageNet-1K	1x	48.3	43.0	72M	google / baidu
ELSA-Swin-S	Mask R-CNN	ImageNet-1K	3x	49.2	43.6	72M	google / baidu
ELSA-Swin-T	Cascade Mask R-CNN	ImageNet-1K	1x	49.8	43.0	86M	google / baidu
ELSA-Swin-T	Cascade Mask R-CNN	ImageNet-1K	3x	51.0	44.2	86M	google / baidu
ELSA-Swin-S	Cascade Mask R-CNN	ImageNet-1K	1x	51.6	44.4	110M	google / baidu
ELSA-Swin-S	Cascade Mask R-CNN	ImageNet-1K	3x	52.3	45.2	110M	google / baidu

ADE20K Semantic Segmentation

Backbone	Method	Pretrain	Crop Size	Lr Schd	mIoU (ms+flip)	#Params	Download
ELSA-Swin-T	UPerNet	ImageNet-1K	512x512	160K	47.9	61M	google / baidu
ELSA-Swin-S	UperNet	ImageNet-1K	512x512	160K	50.4	85M	google / baidu

Install

Clone this repo:

git clone https://github.com/damo-cv/ELSA.git elsa
cd elsa

Create a conda virtual environment and activate it:

conda create -n elsa python=3.7 -y
conda activate elsa

Install PyTorch==1.8.0 and torchvision==0.9.0 with CUDA==10.1:

conda install pytorch==1.8.0 torchvision==0.9.0 cudatoolkit=10.1 -c pytorch

Install CUDA==10.1 with cudnn7 following the official installation instructions
Install Apex:

git clone https://github.com/NVIDIA/apex
cd apex
pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
cd ../

Install mmcv-full==1.3.0

pip install mmcv-full==1.3.0 -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.8.0/index.html

Install other requirements:

pip install -r requirements.txt

Install mmdet and mmseg:

cd ./det
pip install -v -e .
cd ../seg
pip install -v -e .
cd ../

Build the elsa operation:

cd ./cls/models/elsa
python setup.py install
mv build/lib*/* .
cp *.so ../../../det/mmdet/models/backbones/elsa/
cp *.so ../../../seg/mmseg/models/backbones/elsa/
cd ../../../

Data preparation

We use standard ImageNet dataset, you can download it from http://image-net.org/. Please prepare it under the following file structure:

$ tree data
imagenet
├── train
│   ├── class1
│   │   ├── img1.jpeg
│   │   ├── img2.jpeg
│   │   └── ...
│   ├── class2
│   │   ├── img3.jpeg
│   │   └── ...
│   └── ...
└── val
    ├── class1
    │   ├── img4.jpeg
    │   ├── img5.jpeg
    │   └── ...
    ├── class2
    │   ├── img6.jpeg
    │   └── ...
    └── ...

Also, please prepare the COCO and ADE20K datasets following their links. Then, please link them to det/data and seg/data.

Evaluation

ImageNet Classification

Run following scripts to evaluate pre-trained models on the ImageNet-1K:

cd cls

python validate.py <PATH_TO_IMAGENET> --model elsa_swin_tiny --checkpoint <CHECKPOINT_FILE> \
  --no-test-pool --apex-amp --img-size 224 -b 128

python validate.py <PATH_TO_IMAGENET> --model elsa_swin_small --checkpoint <CHECKPOINT_FILE> \
  --no-test-pool --apex-amp --img-size 224 -b 128

python validate.py <PATH_TO_IMAGENET> --model elsa_swin_base --checkpoint <CHECKPOINT_FILE> \
  --no-test-pool --apex-amp --img-size 224 -b 128 --use-ema

COCO Detection

Run following scripts to evaluate a detector on the COCO:

cd det

# single-gpu testing
python tools/test.py <CONFIG_FILE> <DET_CHECKPOINT_FILE> --eval bbox segm

# multi-gpu testing
tools/dist_test.sh <CONFIG_FILE> <DET_CHECKPOINT_FILE> <GPU_NUM> --eval bbox segm

ADE20K Semantic Segmentation

Run following scripts to evaluate a model on the ADE20K:

cd seg

# single-gpu testing
python tools/test.py <CONFIG_FILE> <SEG_CHECKPOINT_FILE> --aug-test --eval mIoU

# multi-gpu testing
tools/dist_test.sh <CONFIG_FILE> <SEG_CHECKPOINT_FILE> <GPU_NUM> --aug-test --eval mIoU

Training from scratch

Due to randomness, the re-training results may have a gap of about 0.1~0.2% with the numbers in the paper.

ImageNet Classification

Run following scripts to train classifiers on the ImageNet-1K:

cd cls

bash ./distributed_train.sh 8 <PATH_TO_IMAGENET> --model elsa_swin_tiny \
  --epochs 300 -b 128 -j 8 --opt adamw --lr 1e-3 --sched cosine --weight-decay 5e-2 \
  --warmup-epochs 20 --warmup-lr 1e-6 --min-lr 1e-5 --drop-path 0.1 --aa rand-m9-mstd0.5-inc1 \
  --mixup 0.8 --cutmix 1. --remode pixel --reprob 0.25 --clip-grad 5. --amp

bash ./distributed_train.sh 8 <PATH_TO_IMAGENET> --model elsa_swin_small \
  --epochs 300 -b 128 -j 8 --opt adamw --lr 1e-3 --sched cosine --weight-decay 5e-2 \
  --warmup-epochs 20 --warmup-lr 1e-6 --min-lr 1e-5 --drop-path 0.3 --aa rand-m9-mstd0.5-inc1 \
  --mixup 0.8 --cutmix 1. --remode pixel --reprob 0.25 --clip-grad 5. --amp

bash ./distributed_train.sh 8 <PATH_TO_IMAGENET> --model elsa_swin_base \
  --epochs 300 -b 128 -j 8 --opt adamw --lr 1e-3 --sched cosine --weight-decay 5e-2 \
  --warmup-epochs 20 --warmup-lr 1e-6 --min-lr 1e-5 --drop-path 0.5 --aa rand-m9-mstd0.5-inc1 \
  --mixup 0.8 --cutmix 1. --remode pixel --reprob 0.25 --clip-grad 5. --amp --model-ema

If GPU memory is not enough when training elsa_swin_base, you can use two nodes (2 * 8 GPUs), each with a batch size of 64 images/GPU.

COCO Detection / ADE20K Semantic Segmentation

Run following scripts to train models on the COCO / ADE20K:

cd det 
# (or cd seg)

# multi-gpu training
tools/dist_train.sh <CONFIG_FILE> <GPU_NUM> --cfg-options model.pretrained=<PRETRAIN_MODEL> [model.backbone.use_checkpoint=True] [other optional arguments]

Acknowledgement

This work was supported by Alibaba Group through Alibaba Research Intern Program and the National Natural Science Foundation of China (No.61976094).

Codebase from pytorch-image-models, ddfnet, VOLO, Swin-Transformer, Swin-Transformer-Detection, and Swin-Transformer-Semantic-Segmentation

Citing ELSA

@article{zhou2021ELSA,
  title={ELSA: Enhanced Local Self-Attention for Vision Transformer},
  author={Zhou, Jingkai and Wang, Pichao and Wang, Fan and Liu, Qiong and Li, Hao and Jin, Rong},
  journal={arXiv preprint arXiv:2112.12786},
  year={2021}
}

You might also like...

The official TensorFlow implementation of the paper Action Transformer: A Self-Attention Model for Short-Time Pose-Based Human Action Recognition

Action Transformer A Self-Attention Model for Short-Time Human Action Recognition This repository contains the official TensorFlow implementation of t

20 Jan 3, 2023

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows" on Object Detection and Instance Segmentation.

Swin Transformer for Object Detection This repo contains the supported code and configuration files to reproduce object detection results of Swin Tran

1.4k Dec 30, 2022

Official implement of "CAT: Cross Attention in Vision Transformer".

CAT: Cross Attention in Vision Transformer This is official implement of "CAT: Cross Attention in Vision Transformer". Abstract Since Transformer has

100 Dec 15, 2022

Implementation of Convolutional enhanced image Transformer

CeiT : Convolutional enhanced image Transformer This is an unofficial PyTorch implementation of Incorporating Convolution Designs into Visual Transfor

82 Dec 13, 2022

Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"

Memory Efficient Attention Pytorch Implementation of a memory efficient multi-head attention as proposed in the paper, Self-attention Does Not Need O(

180 Jan 5, 2023

Implementation of self-attention mechanisms for general purpose. Focused on computer vision modules. Ongoing repository.

Comments

CVE-2007-4559 Patch

Patching CVE-2007-4559

Hi, we are security researchers from the Advanced Research Center at Trellix. We have began a campaign to patch a widespread bug named CVE-2007-4559. CVE-2007-4559 is a 15 year old bug in the Python tarfile package. By using extract() or extractall() on a tarfile object without sanitizing input, a maliciously crafted .tar file could perform a directory path traversal attack. We found at least one unsantized extractall() in your codebase and are providing a patch for you via pull request. The patch essentially checks to see if all tarfile members will be extracted safely and throws an exception otherwise. We encourage you to use this patch or your own solution to secure against CVE-2007-4559. Further technical information about the vulnerability can be found in this blog.

If you have further questions you may contact us through this projects lead researcher Kasimir Schulz.

opened by TrellixVulnTeam 0
Question on the relative positional embedding.

Hi, Thank you for your sharing. When I looked through this repo, I found there is no relative positional embedding or relative bias, which have been mentioned in equations?.

opened by PeiqinZhuang 1

The official implementation of ELSA: Enhanced Local Self-Attention for Vision Transformer

Related tags

Overview

ELSA: Enhanced Local Self-Attention for Vision Transformer

Introduction

Model zoo

ImageNet Classification

COCO Object Detection

ADE20K Semantic Segmentation

Install

Data preparation

Evaluation

ImageNet Classification

COCO Detection

ADE20K Semantic Segmentation

Training from scratch

ImageNet Classification

COCO Detection / ADE20K Semantic Segmentation

Acknowledgement

Citing ELSA

You might also like...

The official TensorFlow implementation of the paper Action Transformer: A Self-Attention Model for Short-Time Pose-Based Human Action Recognition

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows" on Object Detection and Instance Segmentation.

Official implement of "CAT: Cross Attention in Vision Transformer".

Implementation of Convolutional enhanced image Transformer

Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"

Implementation of self-attention mechanisms for general purpose. Focused on computer vision modules. Ongoing repository.

Graph neural network message passing reframed as a Transformer with local attention

Implementation of CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification

Pytorch implementation of ICASSP 2022 paper Attention Probe: Vision Transformer Distillation in the Wild

Comments

CVE-2007-4559 Patch

Patching CVE-2007-4559

Question on the relative positional embedding.

Owner

DamoCV

Official code for "Focal Self-attention for Local-Global Interactions in Vision Transformers"

Implementation of the 😇 Attention layer from the paper, Scaling Local Self-Attention For Parameter Efficient Visual Backbones

Implementation of Deformable Attention in Pytorch from the paper "Vision Transformer with Deformable Attention"

Implementation of Transformer in Transformer, pixel level attention paired with patch level attention for image classification, in Pytorch

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

Official Pytorch Implementation of Relational Self-Attention: What's Missing in Attention for Video Understanding

Official implementation of CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification

Official code for paper "Demystifying Local Vision Transformer: Sparse Connectivity, Weight Sharing, and Dynamic Weight"

Episodic Transformer (E.T.) is a novel attention-based architecture for vision-and-language navigation. E.T. is based on a multimodal transformer that encodes language inputs and the full episode history of visual observations and actions.

This repository builds a basic vision transformer from scratch so that one beginner can understand the theory of vision transformer.