global-and-dual-attention-mechanisms
yolov5改进Enhanced Vehicle Detection in SAR Images via Global and Dual Attention Mechanisms
https://github.com/ynlsj/global-and-dual-attention-mechanisms
Science Score: 54.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (7.4%) to scientific vocabulary
Repository
yolov5改进Enhanced Vehicle Detection in SAR Images via Global and Dual Attention Mechanisms
Basic Info
- Host: GitHub
- Owner: ynlsj
- License: agpl-3.0
- Language: Python
- Default Branch: main
- Homepage: https://github.com/ynlsj/Global-and-Dual-Attention-Mechanisms/
- Size: 20 MB
Statistics
- Stars: 3
- Watchers: 1
- Forks: 1
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Improvements to YOLOv5
Module 1
Refer to $A^2-Nets$，which can be regarded as an evolved version of SE.
Double Attention Method
Module 2
The SPPF of YOLOv5 has been improved. While maintaining the same receptive field, the speed of the model is further enhanced.
SPPF
SPPFCSPC
Module 3
Refer to the improvements to YOLO by others on GitHub,"GAM_Attention" to the feature maps at different levels of the prediction head. This attention method can, to some extent, solve problems such as occlusion and cross - overlap'''
class GAM_Attention(nn.Module):
def __init__(self, in_channels, out_channels, rate=4):
super(GAM_Attention, self).__init__()
self.channel_attention = nn.Sequential(
nn.Linear(in_channels, int(in_channels / rate)),
nn.ReLU(inplace=True),
nn.Linear(int(in_channels / rate), in_channels)
)
self.spatial_attention = nn.Sequential(
nn.Conv2d(in_channels, int(in_channels / rate), kernel_size=7, padding=3),
nn.BatchNorm2d(int(in_channels / rate)),
nn.ReLU(inplace=True),
nn.Conv2d(int(in_channels / rate), out_channels, kernel_size=7, padding=3),
nn.BatchNorm2d(out_channels)
)
def forward(self, x):
b, c, h, w = x.shape
x_permute = x.permute(0, 2, 3, 1).view(b, -1, c)
x_att_permute = self.channel_attention(x_permute).view(b, h, w, c)
x_channel_att = x_att_permute.permute(0, 3, 1, 2)
x = x * x_channel_att
x_spatial_att = self.spatial_attention(x).sigmoid()
out = x * x_spatial_att
return out
'''
Brief Overview of the Above Implementation: Channel Attention: Learn a set of weights through a Multi - Layer Perceptron (MLP) and perform a weighted calculation with the original input. Spatial Attention: Use a Convolutional Neural Network to map the input features weighted by the channels. Map the result to the range of 0 - 1 through the sigmoid activation function, and then perform a weighted calculation with the original input, enabling the model to focus more on the regions of interest.
Overall Improvements
Author Shengjie LEI , Zhiyong WEI , Yulian ZHANG , Meihua FANG , Baowen WU Subject: "Enhanced Vehicle Detection in SAR Images via Global and Dual Attention Mechanisms" The Visual Computer
Owner
- Login: ynlsj
- Kind: user
- Repositories: 1
- Profile: https://github.com/ynlsj
Citation (CITATION.cff)
cff-version: 1.2.0
preferred-citation:
type: software
message: If you use YOLOv5, please cite it as below.
authors:
- family-names: Jocher
given-names: Glenn
orcid: "https://orcid.org/0000-0001-5950-6979"
title: "YOLOv5 by Ultralytics"
version: 7.0
doi: 10.5281/zenodo.3908559
date-released: 2020-5-29
license: AGPL-3.0
url: "https://github.com/ultralytics/yolov5"
GitHub Events
Total
- Watch event: 3
- Push event: 8
- Fork event: 1
- Create event: 3
Last Year
- Watch event: 3
- Push event: 8
- Fork event: 1
- Create event: 3