MJPNet-S*: Multistyle Joint-Perception Network With Knowledge Distillation for Drone RGB-Thermal Crowd Density Estimation in Smart Cities

Crowd density estimation has gained significant research interest owing to its potential in various industries and social applications. Therefore, this article proposes a multistyle joint-perception network based on a knowledge distillation-trained student network (MJPNet-S*) for drone-based red–green–blue, thermal/depth (RGB-T/D) crowd density estimation tasks. To provide superior accuracy and efficiency, a novel trimodal working module effectively combines the modalities to facilitate comprehensive extraction and utilization. A two-step strategy comprising high- and low-level fusion is employed in which the high-level features capture relational reasoning and a 1-D projection relationship module captures multisensory field information with high-quality semantics. A shallow injection fusion module leverages the multiscale and channel relationships at the low level to combine full-text information interactively. Finally, to reduce resource consumption, a neighboring collaborative distillation method enables the lightweight student network to achieve superior performance by increasing the speed by 92% and reducing the number of parameters by 83% compared with those of the teacher. Extensive experiments demonstrate that the proposed MJPNet-S* performs remarkably well on two RGB-T data sets. The code will be made public at
https://github.com/WBangG/MJPNet
.