Computing trajectory similarity is a fundamental task in trajectory analysis. Traditional heuristic methods suffer from quadratic computational complexity, which limits their scalability to large datasets. Recently, Trajectory Representation Learning (TRL) has been extensively studied to address this limitation. However, most existing TRL algorithms face two key challenges. First, they prioritize spatial similarity while neglecting the intricate spatio-temporal dynamics of trajectories, particularly temporal regularities. Second, these methods are often constrained by predefined single spatial or temporal scales, which can significantly impact performance, since the measurement of trajectory similarity depends on spatial and temporal resolution. To address these issues, we propose MSST, a Multi-Scale Self-supervised Trajectory Representation Learning framework. MSST simultaneously processes spatial and temporal information by generating 3D spatial-temporal tokens, thereby capturing spatio-temporal characteristics of trajectories more effectively. Further, MSST explore the multi-scale characteristics of trajectories. Finally, self-supervised contrastive learning is employed to enhance the consistency between the trajectory representations from different views. Experimental results on three real-world datasets for similarity trajectory computation provide insight into the design properties of our approach and demonstrate the superiority of our approach over existing TRL methods. MSST significantly surpasses all state-of-the-art competitors in terms of effectiveness, efficiency, and robustness. We explore the multi-scale characteristics of trajectories. To the best of our knowledge, this is the first effort in the TRL literature. Compared to previous TRL research, the proposed method can balance the noise and the details of trajectories, enabling a more comprehensive analysis by accounting for the variability inherent in trajectory data across different scales.

Trajectory clustering is a cornerstone task in the field of trajectory mining. With the proliferation of deep learning, deep trajectory clustering has been widely researched to mine mobility patterns from massive unlabeled trajectories. Nevertheless, existing methods mostly ignore trajectories' temporal regularities, which are essential for mining fine-grained mobility patterns for applications including traveling group identification, transportation mode discovering, social security emergency, etc. To fill this gap, we propose ConDTC, a contrastive deep trajectory clustering method targeting for fine-grained mobility pattern mining. Specifically, we first design a spatial-temporal trajectory representation learning method which can capture both spatial and temporal regularities of trajectories synchronously. The proposed trajectory representation model can be used as a pre-trained model to serve various downstream trajectory mining tasks. Then, we construct a contrastive trajectory clustering module which optimizes trajectory representations and clustering performance simultaneously. Experimental results on three datasets validate that ConDTC can identify fine-grained mobility patterns by clustering trajectories with similar spatial-temporal mobility patterns together while separating those with different mobility patterns apart. Actually, ConDTC outperforms all state-of-the-art competitors substantially in terms of effectiveness, efficiency and robustness.

Trajectory data often suffer from sparsity and a lack of ground truth data, presenting challenges for map-matching.Traditional map-matching approaches struggle with sparse data, and deep learning methods are hindered by data scarcity.To address these issues, we propose a two-stage map-matching method.In the first stage, we use synthetic trajectory data to pre-train a deep learning model, overcoming data scarcity.We introduce a token mixing scheme during pre-training to clarify the interrelations between GPS-based and segment-based trajectory data.In the second stage, we fine-tune the model with limited ground truth data using a Transformer architecture.Empirical evaluations show that our model achieves 3.69%, 2.21% average precision and recall improvements over the best baseline model with various sparseness.

In the realm of human mobility data analysis, a multitude of constraints result in the publication of sparse, non-uniform implicit trajectories without explicit location information, such as coordinates. Researchers have dedicated substantial efforts towards trajectory recovery, aiming to densify trajectories and gain a more comprehensive understanding of human mobility. However, existing trajectory recovery methods focus on explicit trajectories, and require extensive historical data to capture users' mobility patterns. Nevertheless, implicit trajectories are usually more sparse than explicit trajectories. Addressing these challenges, we propose TrajBERT, an innovative BERT-based trajectory recovery method with spatial-temporal refinement. TrajBERT employs the Transformer encoder to learn mobility patterns bi-directionally and enhances the predictions by cross-stage temporal refinement. Subsequently, we design an output layer with global spatial refinement with a novel spatial-temporal aware loss function. To evaluate the performance of TrajBERT, we conduct a series of experiments on real-world datasets. Remarkably,TrajBERT yields at least 8.2% performance improvement compared to the state-of-the-art trajectory recovery approachs. Furthermore, TrajBERT successfully mitigates the cold start problem commonly experienced with new users lacking historical trajectories. It also shows superior robustness when faced with extremely sparse trajectories, thus demonstrating its potential as a practical tool in the field of human mobility analysis.

Trajectory clustering is a cornerstone task in trajectory mining. Sparse and noisy trajectories like Call Detail Records (CDR) have become popular with the rapid development of mobile applications. However, existing trajectory clustering methods' performance is limited on these trajectories. Therefore, we propose a dual Self-supervised Deep Trajectory Clustering (SDTC) method, to optimize trajectory representation and clustering jointly. First, we leverage the BERT model to learn spatial-temporal mobility patterns and incorporate them into the embeddings of location IDs. Second, we fine-tune the BERT model to learn cluster-friendly representations of trajectories by designing a dual self-supervised cluster layer, which improves the intra-cluster similarities and inter-cluster dissimilarities. Third, we conduct extensive experiments with two real-world datasets. Results show that SDTC improves the clustering accuracy by 12.1% (on a noisy and sparse dataset) and 3.8% (on a very sparse dataset) compared with SOTA deep clustering methods.

Video indexing and retrieval have a wide spectrum of promising applications, motivating the interest of researchers worldwide. This paper offers a tutorial and an overview of the landscape of general strategies in visual content-based video indexing and retrieval, focusing on methods for video structure analysis, including shot boundary detection, key frame extraction and scene segmentation, extraction of features including static key frame features, object features and motion features, video data mining, video annotation, video retrieval including query interfaces, similarity measure and relevance feedback, and video browsing. Finally, we analyze future research directions.

How to fuse static and dynamic information is a key issue in event analysis. In this paper, a top-down motion guided fusing method is proposed for recognizing events in an unconstrained news video. In the method, the static information is represented as a Bag-of-SIFT-features and motion information is employed to generate event specific attention map to direct the sampling of the interest points. We build class-specific motion histograms for each event so as to give more weight on the interest points that are discriminative to the corresponding event. Experimental results on TRECVID 2005 video corpus demonstrate that the proposed method can improve the mean average accuracy of recognition.

Behavior recognition is an attractive direction in the computer vision domain. In this paper, we propose a novel behavior recognition method based on prototype learning using metric learning. Prototype learning algorithm can improve the classification performance of nearest-neighbor classifier, reduce the storage and computation requirements. And the metric learning algorithm is used to advance the performance of the prototype learning. In this paper, We use a kind of compound feature including local feature and motion feature to recognize human behaviors. The experimental results show the effectiveness of our method.

Motion analysis is a very attractive research direction in computer vision field. In this paper, we propose a framework for analyzing real vehicle motion in visual traffic surveillance by using Segment Model (SM), which is a kind of probabilistic model. SM can grasp the underlying information of observation sequence by using segment distribution. It has been proved to be more precise than that of HMM. In the experiments, we compare our approach with the template matching method based on the Hausdorff distance and the state space method based on the Hidden Markov Model (HMM). The experimental results show the effectiveness of our approach.