TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios

Zhu, Xingkui; Lyu, Shuchang; Wang, Xu; Zhao, Qi

doi:10.1109/iccvw54120.2021.00312

Cited by 1,149 publications

(596 citation statements)

References 31 publications

Supporting

Mentioning

330

Contrasting

Order By: Relevance

“…To demonstrate the superiority of the proposed method, a series of experiments are conducted on the P-DSD dataset using several of the best-known algorithms, including Faster R-CNN [26], Cascade R-CNN [28], Double-Head R-CNN [29], YOLOv3 [59], YOLOX [34], YOLOv4 [33], PP-YOLO [60], TPH-YOLOv5 [61]. Among them, PP-YOLO and TPH-YOLOv5 are newer algorithms that perform well on small targets.…”

Section: F Overall Results and Methods Comparisonmentioning

confidence: 99%

Small Vessel Detection Based on Adaptive Dual-Polarimetric Feature Fusion and Sea–Land Segmentation in SAR Images

Zhou

Zhang

et al. 2022

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

View full text Add to dashboard Cite

Detection of small sea vessels in synthetic aperture radar (SAR) images has received much attention in recent years because the small vessels have weak scattering intensity and few image pixels. The existing detection network structures are not well adapted to small-scale targets, the polarimetric data are not properly utilized and the sea-land segmentation process to remove land false alarms is time-consuming. Regarding these problems, firstly, a single low-level path aggregation network (S-LPAN) is designed specifically for small target. The structure reduces false alarms at the feature level by finding suitable single-scale feature maps for detection and adding a semantic enhancement module. Secondly, adaptive dual-polarimetric feature fusion (ADPFF) is proposed to filter the multi-channel features acquired by dual-polarimetric decomposition to reduce feature redundancy. Thirdly, a segmentation layer is added to the network to shield the land from false alarms. The detection and segmentation layers share the feature extraction and feature fusion modules and are jointly trained by a joint loss. Finally, polarimetric SAR detection and segmentation dataset (P-DSD) containing small vessel detection and sea-land segmentation labels is created with reference to the LS-SSDD v1.0 dataset, and experimental results on this dataset verify the improvement of this proposed method over other typical methods.

show abstract

Section: F Overall Results and Methods Comparisonmentioning

confidence: 99%

Small Vessel Detection Based on Adaptive Dual-Polarimetric Feature Fusion and Sea–Land Segmentation in SAR Images

Zhou

Zhang

et al. 2022

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

View full text Add to dashboard Cite

show abstract

“…We implemented the proposed FAFD in Pytorch 1.10.1 environment with NVIDIA RTX 6000 GPU. YOLOv5 consists of five models: YOLOv5n, YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x [44]. Sharing the basic module, these five models have different depths and sizes.…”

Section: Methodsmentioning

confidence: 99%

FAFD: Fast and Accurate Face Detector

Kim

Won

2022

Electronics

View full text Add to dashboard Cite

Deep Neural Networks (DNN) have contributed a significant performance improvement in face detection. However, since most models focus only on the improvement of detection accuracy with computationally expensive structures, it is still far from real-time applications with a fast face detector. The goal of this paper is to improve face detection performance from the speed-focusing point of view. To this end, we propose a novel Fast and Accurate Face Detector (FAFD) to achieve high performance on both speed and accuracy performance. Specifically, based on the YOLOv5 model, we add one prediction head to increase the detection performance, especially for small faces. In addition, to increase the detection performance of multi-scale faces, we propose to add a novel Multi-Scale Image Fusion (MSIF) layer to the backbone network. We also propose an improved Copy-Paste to augment the training images with face objects in various scales. Experimental results on the WiderFace dataset show that the proposed FAFD achieves the best performance among the existing methods in a Speed-Focusing group. On three sub-datasets of WiderFace (i.e., Easy, Medium, and Hard sub-datasets), our FAFD yields average precisions (AP) of 95.0%, 93.5%, and 87.0%, respectively. Also, the speed performance of the FAFD is fast enough to be included in the group of speed-focusing methods.

show abstract

“…There are also certain methods to improve the detection performance of small objects such as designing appropriate anchor size [ 39 , 40 ], introducing visual attention mechanism [ 41 , 42 ], and data augmentation [ 43 , 44 ]. Recently, transformer technology has been gradually applied to computer vision, which provides a new solution to the problem of small object detection [ 45 , 46 ]…”

Section: Related Workmentioning

confidence: 99%

Tiny Vehicle Detection for Mid-to-High Altitude UAV Images Based on Visual Attention and Spatial-Temporal Information

Jiang

et al. 2022

Sensors

View full text Add to dashboard Cite

Mid-to-high altitude Unmanned Aerial Vehicle (UAV) imagery can provide important remote sensing information between satellite and low altitude platforms, and vehicle detection in mid-to-high altitude UAV images plays a crucial role in land monitoring and disaster relief. However, the high background complexity of images and limited pixels of objects challenge the performance of tiny vehicle detection. Traditional methods suffer from poor adaptation ability to complex backgrounds, while deep neural networks (DNNs) have inherent defects in feature extraction of tiny objects with finite pixels. To address the issue above, this paper puts forward a vehicle detection method combining the DNNs-based and traditional methods for mid-to-high altitude UAV images. We first employ the deep segmentation network to exploit the co-occurrence of the road and vehicles, then detect tiny vehicles based on visual attention mechanism with spatial-temporal constraint information. Experimental results show that the proposed method achieves effective detection of tiny vehicles in complex backgrounds. In addition, ablation experiments are performed to inspect the effectiveness of each component, and comparative experiments on tinier objects are carried out to prove the superior generalization performance of our method in detecting vehicles with a limited size of 5 × 5 pixels or less.

show abstract

TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios

Cited by 1,149 publications

References 31 publications

Small Vessel Detection Based on Adaptive Dual-Polarimetric Feature Fusion and Sea–Land Segmentation in SAR Images

Small Vessel Detection Based on Adaptive Dual-Polarimetric Feature Fusion and Sea–Land Segmentation in SAR Images

FAFD: Fast and Accurate Face Detector

Tiny Vehicle Detection for Mid-to-High Altitude UAV Images Based on Visual Attention and Spatial-Temporal Information

Contact Info

Product

Resources

About