In the realm of video analytics, the significance of video anomaly detection cannot be ignored. Many video anomaly detection methods have been introduced in traffic scenarios, and they have achieved some promising results. However, there still exists a large room for improving the traffic anomaly detection performance. In this paper, we present an end-to-end approach, the Dual-Stream Anomaly Detection Network (DS-ADN), specialized for traffic anomaly detection. DS-ADN introduces optical flow data and utilizes inter-flow information interaction coupled with the Multi-Scale Attention Fusion Module (MSA-FM) to comprehensively merge RGB and optical flow data. It features an innovative depth-based prediction loss function that adapts to the vehicle’s regular shape and events at different distances under the surveillance camera. This adaptation occurs through block-wise average pooling and weighted allocation across varying depths. Considering the prevalent issues of datasets dedicated to traffic anomaly detection generally containing a single type of anomaly, this paper introduces the Traffic Surveillance Dataset (TSD) containing multiple anomalies for model validation. Extensive experiments on the TSD dataset highlight the better performance of our DS-ADN, manifesting an AUC gain of over 8.8% compared to other state-of-the-art models. Moreover, DS-ADN achieves competitive performance across three widely-used anomaly detection datasets that contain multiple anomalies, including UCSD Ped2, Avenue, and ShanghaiTech, demonstrating its generalization ability.
Research on vehicle trajectory prediction based on road monitoring video data often utilizes a global map as an input, disregarding the fact that drivers rely on the road structures observable from their own positions for path planning. This oversight reduces the accuracy of prediction. To address this, we propose the CVAE-VGAE model, a novel trajectory prediction approach. Initially, our method transforms global perspective map data into vehicle-centric map data, representing it through a graph structure. Subsequently, Variational Graph Auto-Encoders (VGAEs), an unsupervised learning framework tailored for graph-structured data, are employed to extract road environment features specific to each vehicle’s location from the map data. Finally, a prediction network based on the Conditional Variational Autoencoder (CVAE) structure is designed, which first predicts the driving endpoint and then fits the complete future trajectory. The proposed CVAE-VGAE model integrates a self-attention mechanism into its encoding and decoding modules to infer endpoint intent and incorporate road environment features for precise trajectory prediction. Through a series of ablation experiments, we demonstrate the efficacy of our method in enhancing vehicle trajectory prediction metrics. Furthermore, we compare our model with traditional and frontier approaches, highlighting significant improvements in prediction accuracy.
In practical classification tasks, the sample distribution of the dataset is often unbalanced; for example, this is the case in a dataset that contains a massive quantity of samples with weak labels and for which concrete identification is unavailable. Even in samples with exact labels, the number of samples corresponding to many labels is small, resulting in difficulties in learning the concepts through a small number of labeled samples. In addition, there is always a small interclass variance and a large intraclass variance among categories. Weak labels, few-shot problems, and fine-grained analysis are the key challenges affecting the performance of the classification model. In this paper, we develop a progressive training technique to address the few-shot challenge, along with a weak-label boosting method, by considering all of the weak IDs as negative samples of every predefined ID in order to take full advantage of the more numerous weak-label data. We introduce an instance-aware hard ID mining strategy in the classification loss and then further develop the global and local feature-mapping loss to expand the decision margin. We entered the proposed method into the Kaggle competition, which aims to build an algorithm to identify individual humpback whales in images. With a few other common training tricks, the proposed approach won first place in the competition. All three problems (weak labels, few-shot problems, and fine-grained analysis) exist in the dataset used in the competition. Additionally, we applied our method to CUB-2011 and Cars-196, which are the most widely-used datasets for fine-grained visual categorization tasks, and achieved respective accuracies of 90.1% and 94.9%. This experiment shows that the proposed method achieves the optimal effect compared with other common baselines, and verifies the effectiveness of our method. Our solution has been made available as an open source project.