Ghostformer: A GhostNet-Based Two-Stage Transformer for Small Object Detection

Li, Sijia; Sultonov, Furkat; Tursunboev, Jamshid; Park, Jun-Hyun; Yun, Sangseok; Kang, Jae-Mo

doi:10.3390/s22186939

Cited by 9 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the eigenfeature map and Ghost feature map are combined to obtain the same number of feature maps at a lower computational cost [29]. Compared with current mainstream convolution operations, the Ghost Module has the following advantages [30,31]:…”

Section: Ghostnet Modelmentioning

confidence: 99%

Feature Extraction and Diagnosis of Periodic Transient Impact Faults Based on a Fast Average Kurtogram–GhostNet Method

Jiang,

Zhao,

Zang

et al. 2024

Processes

View full text Add to dashboard Cite

This paper proposes an improved fault diagnosis algorithm that combines a modified fast kurtogram (FK) method with the lightweight convolutional neural network GhostNet. The FK algorithm can adaptively select resonance demodulation bands for envelope demodulation to extract fault features, but it may be disturbed by non-Gaussian noise. Hence, the fast average kurtogram (FAK) method based on sub-band averaging was introduced. This method effectively weakens the impact of pulse noise on the kurtosis graph by splitting the signal into equal-length sub-signals and calculating the average kurtosis value of all sub-signal filters. Simultaneously, to fully utilize the advantages of deep learning technology in feature extraction and classification, this study used the FAK to convert vibration signals from one-dimensional to two-dimensional kurtosis graphs as the input for the GhostNet model. This combination not only achieved accurate fault diagnosis and classification but also showed significant advantages in processing efficiency and resource utilization. The experimental results indicate that the algorithm excelled in extracting features and diagnosing periodic transient impact faults, and compared with traditional methods, it exhibited noticeable improvements in computational efficiency and resource management.

show abstract

Section: Ghostnet Modelmentioning

confidence: 99%

Feature Extraction and Diagnosis of Periodic Transient Impact Faults Based on a Fast Average Kurtogram–GhostNet Method

Jiang,

Zhao,

Zang

et al. 2024

Processes

View full text Add to dashboard Cite

show abstract

“…Another crucial element that impacts feature quality is the backbone network and its ability to extract both semantic and high-resolution features. GhostNet introduced in [93], offers a streamlined and more efficient network that delivers high-quality, multi-scale features to the transformer. Their Ghost module in this network partially generates the output feature map, with the remainder being recovered using simple linear operations.…”

Section: Improved Feature Representationmentioning

confidence: 99%

Extended Expectation Maximization for Under-Fitted Models

Rekavandi¹,

SEGHOUANE²,

Boussaid³

et al. 2022

Preprint

View full text Add to dashboard Cite

<p>In this paper, we generalize the well-known Expectation Maximization (EM) algorithm using the α− divergence for Gaussian Mixture Model (GMM). This approach is used in robust subspace detection when the number of parameters is kept small to avoid overfitting and large estimation variances. The level of robustness can be tuned by the parameterα. When α→1, our method is equivalent to the standard EM approach and for α <1 the method is robust against potential outliers. Simulation results show that the method outperforms the standard EM when it comes to mismatches between noise models and their realizations. In addition, we use the proposed method to detect active brain areas using collected functional Magnetic Resonance Imaging (fMRI) data during task-related experiments.</p>

show abstract

“…But YOLOv4 still has the problem of poor matching of manually designed anchor boxes on different tasks. Due to the excessive number of candidate boxes generated during the prediction process of one-stage methods [15][16][17][18][19][20], non-maximum suppression (NMS) processes are required to filter out a large number of candidate boxes, which not only reduces the inference speed but also fails to achieve truly end-to-end prediction, as shown in [21,22], due to the incorporation of a supervisory mechanism.…”

Section: Introductionmentioning

confidence: 99%

Simple Conditional Spatial Query Mask Deformable Detection Transformer: A Detection Approach for Multi-Style Strokes of Chinese Characters

Zhou,

Xie,

Zhang

et al. 2024

Sensors

View full text Add to dashboard Cite

In the Chinese character writing task performed by robotic arms, the stroke category and position information should be extracted through object detection. Detection algorithms based on predefined anchor frames have difficulty resolving the differences among the many different styles of Chinese character strokes. Deformable detection transformer (deformable DETR) algorithms without predefined anchor frames result in some invalid sampling points with no contribution to the feature update of the current reference point due to the random sampling of sampling points in the deformable attention module. These processes cause a reduction in the speed of the vector learning stroke features in the detection head. In view of this problem, a new detection method for multi-style strokes of Chinese characters, called the simple conditional spatial query mask deformable DETR (SCSQ-MDD), is proposed in this paper. Firstly, a mask prediction layer is jointly determined using the shallow feature map of the Chinese character image and the query vector of the transformer encoder, which is used to filter the points with actual contributions and resample the points without contributions to address the randomness of the correlation calculation among the reference points. Secondly, by separating the content query and spatial query of the transformer decoder, the dependence of the prediction task on the content embedding is relaxed. Finally, the detection model without predefined anchor frames based on the SCSQ-MDD is constructed. Experiments are conducted using a multi-style Chinese character stroke dataset to evaluate the performance of the SCSQ-MDD. The mean average precision (mAP) value is improved by 3.8% and the mean average recall (mAR) value is improved by 1.1% compared with the deformable DETR in the testing stage, illustrating the effectiveness of the proposed method.

show abstract

Ghostformer: A GhostNet-Based Two-Stage Transformer for Small Object Detection

Cited by 9 publications

References 24 publications

Feature Extraction and Diagnosis of Periodic Transient Impact Faults Based on a Fast Average Kurtogram–GhostNet Method

Feature Extraction and Diagnosis of Periodic Transient Impact Faults Based on a Fast Average Kurtogram–GhostNet Method

Extended Expectation Maximization for Under-Fitted Models

Simple Conditional Spatial Query Mask Deformable Detection Transformer: A Detection Approach for Multi-Style Strokes of Chinese Characters

Contact Info

Product

Resources

About