pprp / Awesome Attention Mechanism In Cv
Licence: mit
计算机视觉中用到的注意力模块和其他即插即用模块PyTorch Implementation Collection of Attention Module and Plug&Play Module
Stars: ✭ 54
Programming Languages
python
139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Awesome Attention Mechanism In Cv
Mspeekcollectionviewdelegateimplementation
A custom paging behavior that peeks the previous and next items in a collection view
Stars: ✭ 265 (+390.74%)
Mutual labels: implementation
Structured Self Attention
A Structured Self-attentive Sentence Embedding
Stars: ✭ 459 (+750%)
Mutual labels: attention-model
Attention ocr.pytorch
This repository implements the the encoder and decoder model with attention model for OCR
Stars: ✭ 278 (+414.81%)
Mutual labels: attention-model
Dnc Tensorflow
A TensorFlow implementation of DeepMind's Differential Neural Computers (DNC)
Stars: ✭ 587 (+987.04%)
Mutual labels: implementation
unity-dijkstras-pathfinding
Dijkstra's Pathfinding Algorithm Unity Implementation. (Not being maintained by me, it is just an experiment.)
Stars: ✭ 80 (+48.15%)
Mutual labels: implementation
Sockeye
Sequence-to-sequence framework with a focus on Neural Machine Translation based on Apache MXNet
Stars: ✭ 990 (+1733.33%)
Mutual labels: attention-model
Attention Ocr Chinese Version
Attention OCR Based On Tensorflow
Stars: ✭ 421 (+679.63%)
Mutual labels: attention-model
Transformer
A TensorFlow Implementation of the Transformer: Attention Is All You Need
Stars: ✭ 3,646 (+6651.85%)
Mutual labels: implementation
Mtan
The implementation of "End-to-End Multi-Task Learning with Attention" [CVPR 2019].
Stars: ✭ 364 (+574.07%)
Mutual labels: attention-model
Gpt 2 Pytorch
Simple Text-Generator with OpenAI gpt-2 Pytorch Implementation
Stars: ✭ 618 (+1044.44%)
Mutual labels: implementation
Ml implementation
Implementation of Machine Learning Algorithms
Stars: ✭ 279 (+416.67%)
Mutual labels: implementation
Dcgan Pytorch
PyTorch Implementation of DCGAN trained on the CelebA dataset.
Stars: ✭ 32 (-40.74%)
Mutual labels: implementation
Ethereum Erc721
Non-fungible token implementation for Ethereum-based blockchains.
Stars: ✭ 253 (+368.52%)
Mutual labels: implementation
Nmt Keras
Neural Machine Translation with Keras
Stars: ✭ 501 (+827.78%)
Mutual labels: attention-model
Authenticationservice
Implementations for JWT Authentication in C# and NodeJS.
Stars: ✭ 47 (-12.96%)
Mutual labels: implementation
Reading comprehension tf
Machine Reading Comprehension in Tensorflow
Stars: ✭ 37 (-31.48%)
Mutual labels: attention-model
Text Classification Pytorch
Text classification using deep learning models in Pytorch
Stars: ✭ 683 (+1164.81%)
Mutual labels: attention-model
Awesome-Attention-Mechanism-in-cv
Table of Contents
Introduction
PyTorch实现多种计算机视觉中网络设计中用到的Attention机制,还收集了一些即插即用模块。由于能力有限精力有限,可能很多模块并没有包括进来,有任何的建议或者改进,可以提交issue或者进行PR。
Attention Mechanism
Plug and Play Module
- ACBlock
- Swish、wish Activation
- ASPP Block
- DepthWise Convolution
- Fused Conv & BN
- MixedDepthwise Convolution
- PSP Module
- RFBModule
- SematicEmbbedBlock
- SSH Context Module
- Some other usefull tools such as concate feature map、flatten feature map
- WeightedFeatureFusion:EfficientDet中的FPN用到的fuse方式
- StripPooling:CVPR2020中核心代码StripPooling
- GhostModule: CVPR2020GhostNet的核心模块
- SlimConv: SlimConv3x3
- Context Gating: video classification
- EffNetBlock: EffNet
- ECCV2020 BorderDet: Border aligment module
- CVPR2019 DANet: Dual Attention
- Object Contextual Representation for sematic segmentation: OCRModule
- FPT: 包含Self Transform、Grounding Transform、Rendering Transform
- DOConv: 阿里提出的Depthwise Over-parameterized Convolution
- PyConv: 起源人工智能研究院提出的金字塔卷积
- ULSAM:用于紧凑型CNN的超轻量级子空间注意力模块
- DGC: ECCV 2020用于加速卷积神经网络的动态分组卷积
- DCANet: ECCV 2020 学习卷积神经网络的连接注意力
- PSConv: ECCV 2020 将特征金字塔压缩到紧凑的多尺度卷积层中
- Dynamic Convolution: CVPR2020 动态滤波器卷积(非官方)
- CondConv: Conditionally Parameterized Convolutions for Efficient Inference
Evaluation
基于CIFAR10+ResNet+待测评模块,对模块进行初步测评。测评代码来自于另外一个库:https://github.com/kuangliu/pytorch-cifar/ 实验过程中,不使用预训练权重,进行随机初始化。
模型 | top1 acc | time | params(MB) |
---|---|---|---|
SENet18 | 95.28% | 1:27:50 | 11,260,354 |
ResNet18 | 95.16% | 1:13:03 | 11,173,962 |
ResNet50 | 95.50% | 4:24:38 | 23,520,842 |
ShuffleNetV2 | 91.90% | 1:02:50 | 1,263,854 |
GoogLeNet | 91.90% | 1:02:50 | 6,166,250 |
MobileNetV2 | 92.66% | 2:04:57 | 2,296,922 |
SA-ResNet50 | 89.83% | 2:10:07 | 23,528,758 |
SA-ResNet18 | 95.07% | 1:39:38 | 11,171,394 |
Paper List
SENet 论文: https://arxiv.org/abs/1709.01507 解读:https://zhuanlan.zhihu.com/p/102035721
Contribute
欢迎在issue中提出补充的文章paper和对应code链接。
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].