Skip to content
Open access

Cross-Domain Multi-Semantic Trajectory Anomaly Detection via Kinematic Constraints and Bidirectional Sequential Reconstruction

Oct 2026 · ISPRS International Journal of Geo-Information · Vol 15, pp. 451 · 0 citations · 29 references

TL;DR

Cross-domain experimental results demonstrate that the proposed framework can overcome the Gaussian smoothing trap of generative models and the cross-domain generalization degradation of complex graph networks, and provides a novel solution for highly robust intelligent video surveillance early-warning in the open world.

Abstract

To address the bottlenecks of weak cross-domain generalization, insufficient representation of complex spatial behaviors, and the scarcity of real anomaly labels in the detection of multi-semantic pedestrian trajectory anomalies (e.g., running, loitering, and wrong-way) in public security, this paper proposes an unsupervised detection framework combining kinematic constraints and a Bidirectional Long Short-Term Memory Autoencoder (Bi-LSTM-AE). While deep graph neural networks and attention-based models capture multi-agent interactions, they are frequently prone to spatial weight drift and severe computational overhead when transferred across disparate surveillance layouts. In contrast, our framework focuses on the intrinsic kinematic manifold of individual trajectories, striking an effective trade-off between cross-domain generalizability and linear computational complexity O(L). To this end, we construct a rigorous cross-domain evaluation paradigm, utilizing normal trajectories from a massive public dataset (SDD) to drive the unsupervised training of the model, aiming to fit a generalized human walking manifold unconstrained by specific physical scenes. To tackle the lack of multi-dimensional evaluation benchmarks, we design a multi-semantic anomaly generation strategy based on underlying kinematic constraints, programmatically synthesizing highly realistic challenging test samples on the ground-truth manifold of the target domain (UCY). Cross-domain experimental results demonstrate that the proposed framework can overcome the Gaussian smoothing trap of generative models and the cross-domain generalization degradation of complex graph networks. In highly challenging unseen scenes, the overall detection F1 score is maintained above 0.90; notably, in the detection of “spatial loop anomalies (loitering)” that relies on a global spatio-temporal receptive field, a precise recall rate of 93.76% is achieved. With linear computational complexity, our method provides a novel solution for highly robust intelligent video surveillance early-warning in the open world.

Read PDF

Similar papers

2026

MBSIN: Multi-Branch Spatiotemporal Interaction Network for Pedestrian Trajectory Prediction

Pedestrian trajectory prediction plays a crucial role in intelligent transportation and autonomous driving applications. This task remains challenging due to the complex spatiotemporal correlations involved. Existing methods typically extract spatial and temporal information independently, overlooking their coupling re...

Ziang Wei, Ze Zhou, Wei-Long Liu et al. · 0 citations
Preprint Oct 2026

Trajectory Prediction Method for person under WSN with Hierarchical Envelope Adversarial Domain Adaptation

Pedestrian trajectory prediction holds significant applications across autonomous driving, intelligent surveillance, and smart cities. However, the complexity and unpredictability of pedestrian interactions make this task a persistent challenge. Current models face two common limitations: 1) Consider only a single type...

Wen-Qian Jiang, Qi-Feng Sun, Xiao-Song Gao et al. · 0 citations
Sep 2026

A Spatio-Temporal Graph Network with Informer-TCN Fusion for Vehicle Trajectory Prediction on Highways

A novel hybrid deep learning framework suitable for cloud-based control platforms, providing a foundational algorithmic solution for vehicle-infrastructure cooperative perception and decision-making and suggests potential for integration into intelligent transportation cloud control platforms.

Zi-Yan Liang, Rui Yuan, Peng-Ying Zhou et al. · 0 citations
Preprint Oct 2026

A Unified Score Matching Paradigm for Video Anomaly Detection and Anticipation

Video anomaly detection (VAD) is a fundamental and safety-critical task in computer vision. Recent generative approaches detect anomalies from a distributional perspective, but remain limited by local anomaly modes. Meanwhile, video anomaly anticipation (VAA), as a proactive extension beyond post-hoc detection, introdu...

Cong-Qi Cao, Zhen Liang, Han-Wen Zhang et al. · 0 citations
Open access Sep 2026

Learn the interactions: Weakly supervised video anomaly detection with human-object interactions

A novel weakly supervised cross-domain framework that enriches feature representations by explicitly integrating semantic cues from Human-Object Interactions (HOI) and action dynamics and achieves competitive performance against state-of-the-art methods under weak supervision in terms of AUC.

Mao-Wen Zhou, Erma Rahayu Mohd Faizal Abdullah, Aznul Qalid Md Sabri et al. · 0 citations
#machine learning Preprint Sep 2026

An Adversarial Zero-Shot Learning Approach for Anomaly Detection in Multivariate IoT Traffic Data

This paper proposes a novel framework for multivariate time-series anomaly detection that leverages adversarial learning and contrastive loss within a sequence-based Variational Autoencoder (VAE) architecture, and introduces encoder and decoder adaptor layers that align feature distributions across domains while preser...

Mahshid Rezakhani, Tolunay Seyfi, Fatemeh Afghah · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.