Crowd counting via an inverse attention residual network

Liu, Yanbo; Jia, Rui-Sheng; Liu, Qingming; Xu, Zhi-Feng; Sun, Hong-Mei

doi:10.1117/1.jei.29.3.033010

Cited by 2 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, due to the individual behaviors of different pedestrians, the normal self-attention mechanism is difficult to produce diverse attention results. In recent years, some works use the inverse method to obtain extra information discarded by the traditional attention model, e.g., pedestrian counting (Si and Patel 2019;Liu et al 2020). Inspired by them, we use the complementary method to obtain both normal and inverse attention.…”

Section: Related Workmentioning

confidence: 99%

Complementary Attention Gated Network for Pedestrian Trajectory Prediction

Duan

Wang

Long³

et al. 2022

AAAI

View full text Add to dashboard Cite

Pedestrian trajectory prediction is crucial in many practical applications due to the diversity of pedestrian movements, such as social interactions and individual motion behaviors. With similar observable trajectories and social environments, different pedestrians may make completely different future decisions. However, most existing methods only focus on the frequent modal of the trajectory and thus are difficult to generalize to the peculiar scenario, which leads to the decline of the multimodal fitting ability when facing similar scenarios. In this paper, we propose a complementary attention gated network (CAGN) for pedestrian trajectory prediction, in which a dual-path architecture including normal and inverse attention is proposed to capture both frequent and peculiar modals in spatial and temporal patterns, respectively. Specifically, a complementary block is proposed to guide normal and inverse attention, which are then be summed with learnable weights to get attention features by a gated network. Finally, multiple trajectory distributions are estimated based on the fused spatio-temporal attention features due to the multimodality of future trajectory. Experimental results on benchmark datasets, i.e., the ETH, and the UCY, demonstrate that our method outperforms state-of-the-art methods by 13.8% in Average Displacement Error (ADE) and 10.4% in Final Displacement Error (FDE). Code will be available at https://github.com/jinghaiD/CAGN

show abstract

Section: Related Workmentioning

confidence: 99%