QueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detection

Yang, Chenhongyi; Huang, Zehao; Wang, Naiyan

doi:10.48550/arxiv.2103.09136

Cited by 8 publications

(5 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, with the threshold set to 0.5:0.95, the mAP value of the improved YOLOv5 algorithm achieves 34.0%, surpassing all other models. DMNet [38] 47.6% 28.2% YOLOv3 [39] 36.6% 17.5% Cascade-RCNN [40] 45.9% 24.3% YOLOv5l 46.4% 28.1% QueryDet [41] 48.2% 28.3% ClusDet [42] 53.2% 30.4% PRNet [43] 53.9% 32.0% Model of this paper 52.5% 34.0%…”

Section: Comparative Experimentsmentioning

confidence: 96%

An Aerial Image Detection Algorithm Based on Improved YOLOv5

Shan,

Yang,

Wang

et al. 2024

Sensors

View full text Add to dashboard Cite

To enhance aerial image detection in complex environments characterized by multiple small targets and mutual occlusion, we propose an aerial target detection algorithm based on an improved version of YOLOv5 in this paper. Firstly, we employ an improved Mosaic algorithm to address redundant boundaries arising from varying image scales and to augment the training sample size, thereby enhancing detection accuracy. Secondly, we integrate the constructed hybrid attention module into the backbone network to enhance the model’s capability in extracting pertinent feature information. Subsequently, we incorporate feature fusion layer 7 and P2 fusion into the neck network, leading to a notable enhancement in the model’s capability to detect small targets. Finally, we replace the original PAN + FPN network structure with the optimized BiFPN (Bidirectional Feature Pyramid Network) to enable the model to preserve deeper semantic information, thereby enhancing detection capabilities for dense objects. Experimental results indicate a substantial improvement in both the detection accuracy and speed of the enhanced algorithm compared to its original version. It is noteworthy that the enhanced algorithm exhibits a markedly improved detection performance for aerial images, particularly under real-time conditions.

show abstract

Section: Comparative Experimentsmentioning

confidence: 96%

An Aerial Image Detection Algorithm Based on Improved YOLOv5

Shan,

Yang,

Wang

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…In the above subsection, the effectiveness of the proposed FFCLC has been verified, now, we choose some representative algorithms to compare with the FFCLC in different datasets of SAR-Ship-Dataset and SSDD. The detection results of Faster R-CNN, SSD, YOLOv3, TPH-YOLOv5 [49], QueryDet [50], YOLOv8, YOLOv5, and FFCLC in four different scenarios of SAR-Ship-Dataset are given in Fig. 12, and Fig.…”

Section: E Comparative Experimentsmentioning

confidence: 99%

A Fast and Accurate Small Target Detection Algorithm Based on Feature Fusion and Cross-Layer Connection Network for the SAR Images

Sun,

Li,

Chen

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Target detection technology has been greatly improved for the synthetic aperture radar (SAR) images recently, due to the advancement in deep learning (DL) domain. However, because of the existence of clutter in the SAR images, it's still a challenge to detect small targets with high accuracy and low computational complexity. To solve this problem, a detection algorithm based on feature fusion and cross-layer connection (FFCLC) network is proposed in this paper. Firstly, the attention feature fusion (AFF) is applied to improve the feature fusion ability for the small targets through allocating weights to various feature maps adaptively. Meanwhile, the depthwise separable convolution (DW-Conv) is used to reduce the computational complexity caused by the increasement of network layers. Then, a cross-layer connection (Cross-Connect) submodule is proposed to fuse shallow features with deep features further. Finally, a multi-scale target detection (Multi-Detect) submodule is designed to improve the detection ability for the small targets. We compare the proposed algorithm with the other representative methods on the SAR-Ship-Dataset and SSDD, quantitative evaluations show that our proposed algorithm's can reach the highest computational efficiency. Therefore, because of the superior performance in terms of accuracy and efficiency, the algorithm proposed in this paper is more suitable to detect small targets for the SAR images.

show abstract

“…In the commonly adopted implementation of feature pyramid-based dense detectors, the predictions of different layers are "flatten and concatenated" for loss computation, making the supervision on different layers to be conjectural. However, as pointed out in [49], the sample distribution is significantly imbalanced between different layers, because the feature size grows up in quadratic as the feature resolution increases. As a result, the training samples will be dominated by samples in lower levels, making the high-level samples lack of supervision, which will harm the performance on large objects.…”

Section: Supervision Conjunction In Fpnmentioning

confidence: 99%

“…VGG-16 0.967 0.959 0.912 AInnoFace [53] ResNet-152 0.970 0.961 0.918 RetinaFace [8] ResNet-152 0.969 0.961 0.918 RefineFace [54] ResNet-152 0.972 0.962 0.920 DSFD [18] ResNet-152 0.966 0.957 0.904 ASFD-D6 [52] ResNet-152 0.972 0.965 0.925 HAMBox [25] ResNet-50 0.970 0.964 0.933 TinaFace [62] ResNet 7. Firstly, we test the Linear-Reweight proposed in [49], in which the factors that linearly increase from 1.0 to 2.0 are assigned to each FPN level. Secondly, considering that most negative samples are greatly suppressed by the focal loss, we propose a vanilla re-weight strategy where the weight of each FPN layer is proportionate to the number of positive samples, which is denoted as Sum-Reweight.…”

Section: What Is the Best Label Assignment Disentanglement?mentioning

confidence: 99%

Disentangle Your Dense Object Detector

Chen

Yang

et al. 2021

Proceedings of the 29th ACM International Conference on Multimedia

Self Cite

101

View full text Add to dashboard Cite

Deep learning-based dense object detectors have achieved great success in the past few years and have been applied to numerous multimedia applications such as video understanding. However, the current training pipeline for dense detectors is compromised to lots of conjunctions that may not hold. In this paper, we investigate three such important conjunctions: 1) only samples assigned as positive in classification head are used to train the regression head; 2) classification and regression share the same input feature and computational fields defined by the parallel head architecture; and 3) samples distributed in different feature pyramid layers are treated equally when computing the loss. We first carry out a series of pilot experiments to show disentangling such conjunctions can lead to persistent performance improvement. Then, based on these findings, we propose Disentangled Dense Object Detector (DDOD), in which simple and effective disentanglement mechanisms are designed and integrated into the current state-of-the-art dense object detectors. Extensive experiments on MS COCO benchmark show that our approach can lead to 2.0 mAP, 2.4 mAP and 2.2 mAP absolute improvements on RetinaNet, FCOS, and ATSS baselines with negligible extra overhead. Notably, our best model reaches 55.0 mAP on the COCO test-dev set and 93.5 AP on the hard subset of WIDER FACE, achieving new state-of-the-art performance on these two competitive benchmarks. Code is available at https://github.com/zehuichen123/DDOD. CCS CONCEPTS• Computing methodologies → Object detection.

show abstract

QueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detection

Cited by 8 publications

References 59 publications

An Aerial Image Detection Algorithm Based on Improved YOLOv5

An Aerial Image Detection Algorithm Based on Improved YOLOv5

A Fast and Accurate Small Target Detection Algorithm Based on Feature Fusion and Cross-Layer Connection Network for the SAR Images

Disentangle Your Dense Object Detector

Contact Info

Product

Resources

About