VisDrone-CC2021: The Vision Meets Drone Crowd Counting Challenge Results

Liu, Zhihao; He, Zhijian; Wang, Lujia; Yang, Yi; Yuan, Yixuan; Zhang, Dingwen; Zhang, Jinglin; Zhu, Pengfei; Gool, Luc Van; Han, Junwei; Hoi, Steven C. H.; Hu, Qinghua; Liu, Ming; Pan, Junwen; Yin, Baocai; Zhang, Binyu; Liu, Chengxin; Ding, Ding; Liang, Dingkang; Ding, Guanchen; Lü, Hao; Lin, Huei-Yung; Chen, Jingyuan; Li, Jiong; Liu, Liang; Lin, Zhenzhi; Shi, Min; Yang, Qianqian; He, Qing; Peng, Sifan; Xu, Wei; Han, Wenwei; Bai, Xiang; Chen, Xiwu; Wang, Yabin; Xia, Yinfeng; Tao, Yiran; Chen, Zhenzhong; Cao, Zhiguo

doi:10.1109/iccvw54120.2021.00317

Cited by 20 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with the best result of Bayesian Loss [70], our model has improved MAE and MSE by 0.54 and 0.53 on the RGBT-CC dataset. Similarly, our method is compared with multiple best-performing multimodal fusion models of UCNet [66], HDFNet [67], and BBSNet [68], and compared with the best-performing BBSNet [7], it is found that our model has 1.4 and 0.34 improvement in MAE and MSE on the RGBT-CC dataset. In the literature of the Sun Yat-Sen University team, the integration of the IADM "early fusion" mechanism into the classical counting model networks such as MCNN [33], SANet [69], CSRNet [33], and Bayesian Loss [70] can improve the performance of the model.…”

Section: Density Regression Methodmentioning

confidence: 98%

Research on 24-Hour Dense Crowd Counting and Object Detection System Based on Multimodal Image Optimization Feature Fusion

Ren

2022

Scientific Programming

View full text Add to dashboard Cite

Motivation. In the environment of day and night video surveillance, in order to improve the accuracy of machine vision dense crowd counting and target detection, this paper designs a day and night dual-purpose crowd counting and crowd detection network based on multimode image fusion. Methods. Two sub-models, RGBD-Net and RGBT-Net, are designed in this paper. The depth image features and thermal imaging features are effectively fused with the features of visible light images, so that the model has stronger anti-interference characteristics and robustness to the light noise interference caused by the sudden fall of light at night. The above models use density map regression-guided detection method to complete population counting and detection. Results. The model completed daytime training and testing on MICC dataset. Through verification, the average absolute error of the model was 1.025, the mean square error was 1.521, and the recall rate of target detection was 97.11%. Night vision training and testing were completed on the RGBT-CC dataset. After verification, the average absolute error of the network was 18.16, the mean square error was 32.14, and the recall rate of target detection was 97.65%. By verifying the effectiveness of the multimode medium-term fusion network, it is found to exceed the current most advanced bimodal fusion method. Conclusion. The experimental results show that the proposed multimodal fusion network can solve the counting and detection problem in the video surveillance environment during day and night. The ablation experiment further proves the effectiveness of the parameters of the two models.

show abstract

Section: Density Regression Methodmentioning

confidence: 98%

Research on 24-Hour Dense Crowd Counting and Object Detection System Based on Multimodal Image Optimization Feature Fusion

Ren

2022

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…How to fine-tune deep neural architectures to achieve an optimal balance between precision and performance is an active research area. The VisDrone Crowd Counting challenge was introduced to encourage research in this direction [8,9]; nevertheless, the solutions proposed by the participants in the challenge are not always focused on efficiency but rather on effectiveness, as the goal is only to obtain a low error in counting people. The lowest error was obtained with TransCrowd [18], based on the increasingly popular Vision Transformer [19].…”

Section: Related Workmentioning

confidence: 99%

“…This was done in our previous preliminary work [7] but, while effective, this approach proved too demanding from a computational point of view. The method was tested on the recently proposed Crowd Counting 2020 [8] and 2021 [9] datasets, used annually for the international VisDrone challenge. The peculiarity of these datasets is that they are not characterized by still images but actually by frames of video sequences that are used here to perform the crowd flow detection task.…”

Section: Introductionmentioning

confidence: 99%

Density-based clustering with fully-convolutional networks for crowd flow detection from drones

Castellano,

Cotardo,

Mencar

et al. 2023

Preprint

View full text Add to dashboard Cite

Crowd analysis from drones has attracted increasing attention in recent times due to the ease of use and affordable cost of these devices. However, how this technology can provide a solution to crowd flow detection is still an unexplored research question. To this end, we propose a crowd flow detection method for video sequences shot by a drone. The method is based on a fully-convolutional network that learns to perform crowd clustering in order to detect the centroids of crowd-dense areas and track their movement in consecutive frames. The proposed method proved effective and efficient when tested on the Crowd Counting datasets of the VisDrone challenge, characterized by video sequences rather than still images. The encouraging results show that the proposed method could open up new ways of analyzing high-level crowd behavior from drones.

show abstract

“…The data set we used is VisDrone2021-counting [ 26 ], from the 2021 Vision Meets Drones: A Challenge. The data set is divided into two parts: train- and test-challenge, including 1807 and 912 RGB images, respectively.…”

Section: Experiments and Evaluationmentioning

confidence: 99%

Improved YOLOv4 for Pedestrian Detection and Counting in UAV Images

Kong

Chen

Yue

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

UAV (unmanned aerial vehicle) captured images have small pedestrian targets and loss of key information after multiple down sampling, which are difficult to overcome by existing methods. We propose an improved YOLOv4 model for pedestrian detection and counting in UAV images, named YOLO-CC. We used the lightweight YOLOv4 for pedestrian detection, which replaces the backbone with CSPDarknet-34, and two feature layers are fused by FPN (Feature Pyramid Networks). We expanded the perception field using multiscale convolution based on the high-level feature map and generated the population density map by feature dimension reduction. By embedding the density map generation method into the network for end-to-end training, our model can effectively improve the accuracy of detection and counting and make feature extraction more focused on small targets. Our experiments demonstrate that YOLO-CC achieves 21.76 points AP50 higher than that of the original YOLOv4 on the VisDrone2021-counting data set while running faster than the original YOLOv4.

show abstract

VisDrone-CC2021: The Vision Meets Drone Crowd Counting Challenge Results

Cited by 20 publications

References 21 publications

Research on 24-Hour Dense Crowd Counting and Object Detection System Based on Multimodal Image Optimization Feature Fusion

Research on 24-Hour Dense Crowd Counting and Object Detection System Based on Multimodal Image Optimization Feature Fusion

Density-based clustering with fully-convolutional networks for crowd flow detection from drones

Improved YOLOv4 for Pedestrian Detection and Counting in UAV Images

Contact Info

Product

Resources

About