2021
DOI: 10.1109/jstars.2021.3122152
|View full text |Cite
|
Sign up to set email alerts
|

Unsupervised Cluster Guided Object Detection in Aerial Images

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
12
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
1
1

Relationship

0
9

Authors

Journals

citations
Cited by 27 publications
(12 citation statements)
references
References 27 publications
0
12
0
Order By: Relevance
“…Additionally, certain methods focus on resolving the issue of drastic scale variations in UAV images by incorporating multi-scale feature fusion [21,22]. Another category of methods adopts a two-stage detection strategy that guides the network to learn focused regions first and then refines the processing of these areas [23,24]. This approach effectively addresses challenges related to the detection of small objects and object clustering in UAV imagery.…”
Section: Uav Detection Methodsmentioning
confidence: 99%
“…Additionally, certain methods focus on resolving the issue of drastic scale variations in UAV images by incorporating multi-scale feature fusion [21,22]. Another category of methods adopts a two-stage detection strategy that guides the network to learn focused regions first and then refines the processing of these areas [23,24]. This approach effectively addresses challenges related to the detection of small objects and object clustering in UAV imagery.…”
Section: Uav Detection Methodsmentioning
confidence: 99%
“…Previous improvement methods for target detection in UAV aerial images can be categorized into three types: (i) utilizing more shallow feature information, such as adding small target detection layers [19]; (ii) enhancing the feature extraction capability of the target detection network, such as improving the Neck network [20] or introducing attention mechanisms [21]; and (iii) increasing input feature information, such as generating higher resolution images [22], image copying [23], and image cropping [24,25].…”
Section: Introductionmentioning
confidence: 99%
“…The UGGNet model proposed by Liao et al uses a local localization LLM to predict the distribution of targets, and then generates target dense regions using an unsupervised clustering module for detection. This method saves a lot of time for detecting small targets [11]. Singh et al proposed a major improved algorithm based on YOLOv5, which can capture small details in the feature map by adding a new feature fusion layer with a small Receptive field in the feature pyramid of YOLOv5.…”
Section: Introductionmentioning
confidence: 99%