Scale Match for Tiny Person Detection

Yu, Xuehui; Gong, Yuqi; Jiang, Nan; Ye, Qixiang; Han, Zhenjun

doi:10.1109/wacv45572.2020.9093394

Cited by 199 publications

(186 citation statements)

References 30 publications

Supporting

Mentioning

155

Contrasting

Order By: Relevance

“…This method not only effectively enhances the detection of small objects, but also promotes the detection of large and medium objects. Yu et al have proposed a scale matching method (Scale Match), which adjusts the image scale of training data and significantly improved the small object detection [21]. Wang et al believe that the commonly used top-down feature pyramid only focuses on transferring the high-level semantics from the top to the bottom, but do not pay attention to transferring the bottom details to the top.…”

Section: A Small Object Detectionmentioning

confidence: 99%

Semi-Supervised Dim and Small Infrared Ship Detection Network Based on Haar Wavelet

et al. 2021

View full text Add to dashboard Cite

Traditional deep learning detection network has poor effect on the detection of infrared dim and small targets on the sea in the case of interference or bad weather. In this paper, an improved dim and small infrared ship detection network based on Haar wavelet is proposed. The HaarConv module is designed based on the high-frequency features obtained by Haar wavelet decomposition, which further increases the feature extraction ability of the backbone network for small targets. Meanwhile, the HaarUp-HaarDown module is designed by using Haar forward and inverse transform, replacing the up-sampling layer of the feature pyramid network and the down-sampling layer of backbone network to retain smaller target features. Furthermore, the pseudo-label-based method enables the network to conduct semisupervised learning, which reduces the labeling cost and improves detection accuracy while expanding the amount of training data. The above method is applied to the YOLOv5-s lightweight network and 11278 infrared images (3352 labeled) of dim and small ships are collected as a dataset. The results show that the introduction of semi-supervised training method effectively expands the training dataset, and the mAP@.5:.95 increases by 23.5%. The proposed Haar wavelet improvement method can effectively improve the detection accuracy of dim and small infrared ship targets by more than 2%, and the number of parameters increases by only about 0.02M. Compared with existing methods, the proposed method reaches the state-of-the-art result and has good generalization performance. INDEX TERMS Haar wavelet, feature map enhancement, dim and small infrared ship, self-training

show abstract

Section: A Small Object Detectionmentioning

confidence: 99%

Semi-Supervised Dim and Small Infrared Ship Detection Network Based on Haar Wavelet

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In images with a resolution of 512 × 512 × 3, the minimum size of a cyst object is approximate 21 × 21, while the minimum size of a nauplius object is around 23 × 23. According to the criteria described in the MS COCO dataset (Lin et al, 2014) and the dataset presented in (Yu, Gong, Jiang, Ye, & Han, 2020), such objects can be categorized into small objects or tiny objects.…”

Section: The Artemia Marker Datasetmentioning

confidence: 99%

Automated detection and counting of Artemia using U-shaped fully convolutional networks and deep convolutional networks

Wang

Stappen²,

Baets

2021

Expert Systems with Applications

View full text Add to dashboard Cite

“…finding missing people and person identification), people counting in a dense crowd, abnormal event detection, public safety, traffic control, and many more (Santos, Disney, and Chave 2018;Sirmacek and Reinartz 2013). However, the task of human detection for UAVs is more complex and demanding due to challenging conditions, such as humans being in small size (Yu et al 2020) due to high altitudes, changing viewing angles, extensive diversity in background, differences in person poses, clothing and appearance, illumination variations, moving platform, and many more (Rudol and Doherty 2008). In addition, partial or complete occlusion caused by distractors (such as vegetation and buildings) adds to these challenges.…”

Section: Introductionmentioning

confidence: 99%

Efficient Real-Time Human Detection Using Unmanned Aerial Vehicles Optical Imagery

Golkarnarenji

Martinez‐Alpiste

Wang

et al. 2021

International Journal of Remote Sensing

View full text Add to dashboard Cite

Unmanned Aerial Vehicles (UAVs) are promising technologies within many different application scenarios including human detection in search and rescue and surveillance use cases, which have received considerable attention worldwide. However, adverse conditions, such as varying altitude, overhead camera placement, changing illumination and moving platform, impose challenges for highperformance yet cost-efficient human detection. To overcome these challenges, we propose a novel combination of dilated convolutions with Path Aggregation Network (PAN) as a new deep neural network-based human detection algorithm in real time. Furthermore, we establish a comprehensive human detection dataset with varying backgrounds, illuminations, and contrast and train the proposed machine-learning model on the collected dataset. Our approach achieves both high precision (88.0% mean Average Precision (mAP)) and real time (67.0 Frames Per Second (FPS)) on a commercial off-the-shelf PC platform. In terms of accuracy, the result is comparable to the standard You Only Look Once v3 (YOLOv3). However, the speed is twice as that of the standard YOLOv3. YOLOv4 is slightly more accurate (89.8%) than our approach. However, it is slower (38.0 versus 67.0 FPS) and has more Billion Floating-Point Operations (BFLOPS). The proposed algorithm has also trained with the VisDrone2019 dataset and compared with seven studies using this dataset. The results have further validated the effectiveness of the proposed approach. Moreover, the algorithm has been evaluated on an embedded system (Jetson AGX Xavier), which demonstrates the usefulness of this method on power-constrained devices. The proposed algorithm is fast, memory efficient, and computationally less expensive to achieve high detection performance. It is expected to contribute significantly to the wider use of UAV applications including search and rescue missions to locate missing people, and surveillance particularly for applications running on resource-constrained platforms, like smartphones or tablets. This proposed system is now being used in aerial drone system of Police of Scotland to help them locate and find missing and vulnerable people. The results of the project were broadcasted by BBC Scotland.

show abstract

Scale Match for Tiny Person Detection

Cited by 199 publications

References 30 publications

Semi-Supervised Dim and Small Infrared Ship Detection Network Based on Haar Wavelet

Semi-Supervised Dim and Small Infrared Ship Detection Network Based on Haar Wavelet

Automated detection and counting of Artemia using U-shaped fully convolutional networks and deep convolutional networks

Efficient Real-Time Human Detection Using Unmanned Aerial Vehicles Optical Imagery

Contact Info

Product

Resources

About