期刊
IEEE TRANSACTIONS ON IMAGE PROCESSING
卷 31, 期 -, 页码 6032-6047出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIP.2022.3205210
关键词
Head; Location awareness; Task analysis; Feature extraction; Annotations; Convolutional neural networks; Context modeling; Video crowd analysis; Gaussian neighborhood attention; VSCrowd dataset; spatial-temporal modeling
The paper proposes a new method for video crowd localization, which models the spatial-temporal dependencies of human mobility using multi-focus Gaussian neighborhood attention. They develop a unified neural network to accurately locate head centers in video clips. They also introduce a large-scale crowd video benchmark for future research in this field.
Video crowd localization is a crucial yet challenging task, which aims to estimate exact locations of human heads in the given crowded videos. To model spatial-temporal dependencies of human mobility, we propose a multi-focus Gaussian neighborhood attention (GNA), which can effectively exploit long-range correspondences while maintaining the spatial topological structure of the input videos. In particular, our GNA can also capture the scale variation of human heads well using the equipped multi-focus mechanism. Based on the multi-focus GNA, we develop a unified neural network called GNANet to accurately locate head centers in video clips by fully aggregating spatial-temporal information via a scene modeling module and a context cross-attention module. Moreover, to facilitate future researches in this field, we introduce a large-scale crowd video benchmark named VSCrowd (https://github.com/HopLee6/VSCrowd), which consists of 60K+ frames captured in various surveillance scenes and 2M+ head annotations. Finally, we conduct extensive experiments on three datasets including our VSCrowd, and the experiment results show that the proposed method is capable to achieve state-of-the-art performance for both video crowd localization and counting.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据