A multiscale sequence information fusion (MSSIF) method is presented for dynamic facial expression recognition (DFER) in video sequences. It exploits multiscale information by integrating features from individual frames, subsequences, and entire sequences through a transformer-based architecture. This hierarchical feature fusion process includes deep feature extraction at the frame level to capture intricate visual details, intrasubsequence fusion using self-attention mechanisms for analyzing adjacent frames, and intersubsequence fusion to synthesize long-term emotional dynamics across time scales. The efficacy of MSSIF is demonstrated through extensive evaluation on three video datasets: eNTERFACE'05, BAUM-1s, and AFEW, where it achieves overall recognition accuracies of 60.1%, 60.7%, and 58.8%, respectively. These results substantiate MSSIF's superior performance in accurately recognizing facial expressions by managing short and long-term dependencies within video sequences, making it a potent tool for real-world applications requiring nuanced dynamic facial expression detection.

The advances in mobile computing and social networking services enable people to probe the dynamics of a city. In this paper, we address the problem of detecting and describing traffic anomalies using crowd sensing with two forms of data, human mobility and social media. Traffic anomalies are caused by accidents, control, protests, sport events, celebrations, disasters and other events. Unlike existing traffic-anomaly-detection methods, we identify anomalies according to drivers' routing behavior on an urban road network. Here, a detected anomaly is represented by a sub-graph of a road network where drivers' routing behaviors significantly differ from their original patterns. We then try to describe the detected anomaly by mining representative terms from the social media that people posted when the anomaly happened. The system for detecting such traffic anomalies can benefit both drivers and transportation authorities, e.g., by notifying drivers approaching an anomaly and suggesting alternative routes, as well as supporting traffic jam diagnosis and dispersal. We evaluate our system with a GPS trajectory dataset generated by over 30,000 taxicabs over a period of 3 months in Beijing, and a dataset of tweets collected from WeiBo, a Twitter-like social site in China. The results demonstrate the effectiveness and efficiency of our system.

The advances in sensor technologies enable real-time collection of high-fidelity spatiotemporal data on transportation networks of major cities. In this paper, using two real-world transportation datasets: 1) incident data and 2) traffic data, we address the problem of predicting and quantifying the impact of traffic incidents. Traffic incidents include any non-recurring events on road networks, including accidents, weather hazard, road construction or work zone closures. By analyzing archived incident data, we classify incidents based on their features (e.g., time, location, type of incident). Subsequently, we model the impact of each incident class on its surrounding traffic by analyzing the archived traffic data at the time and location of the incidents. Consequently, in real-time, if we observe a similar incident (from real-time incident data), we can predict and quantify its impact on the surrounding traffic using our developed models. This information, in turn, can help drivers to effectively avoid impacted areas in real-time. To be useful for such real-time navigation application, and unlike current approaches, we study the dynamic behavior of incidents and model the impact as a quantitative time varying spatial span. In addition to utilizing incident features, we improve our classification approach further by analyzing traffic density around the incident area and the initial behavior of the incident. We evaluated our approach with very large traffic and incident datasets collected from the road networks of Los Angeles County and the results show that we can improve our baseline approach, which solely relies on incident features, by up to 45%.

For the first time, real-time high-fidelity spatiotemporal data on transportation networks of major cities have become available. This gold mine of data can be utilized to learn about traffic behavior at different times and locations, potentially resulting in major savings in time and fuel, the two important commodities of 21st century. As a first step towards the utilization of this data, in this paper, we study the real-world data collected from Los Angeles County transportation network in order to incorporate the data's intrinsic behavior into a time-series mining technique to enhance its accuracy for traffic prediction. In particular, we utilized the spatiotemporal behaviors of rush hours and events to perform a more accurate prediction of both short-term and long-term average speed on road-segments, even in the presence of infrequent events (e.g., accidents). Our result shows that taking historical rush-hour behavior we can improve the accuracy of traditional predictors by up to 67% and 78% in short-term and long-term predictions, respectively. Moreover, we can incorporate the impact of an accident to improve the prediction accuracy by up to 91%.

As geo-realistic rendering of land surfaces is becoming commonplace in geographical information systems (GIS), games and online Earth visualization platforms, a new type of k Nearest Neighbor (kNN) queries, "surface" k Nearest Neighbor (skNN) queries, has emerged and been investigated recently, which extends the traditional kNN queries to a constrained third dimension (i.e., land surface). All existing techniques, however, assume a static environment, limiting their utility in emerging applications (e.g., Location-based Services) where objects move. In this paper, for the first time, we propose two exact methods that can continuously answer skNN queries in a highly dynamic environment which allows for arbitrary movements of data objects. The first method, inspired by the existing techniques in monitoring kNN in road networks [7] maintains an analogous counterpart of the Dijkstra Expansion Tree on land surface, called Surface Expansion Tree (SE-Tree). However, we show the concept of expansion tree for land surface does not work as SE-tree suffers from intrinsic defects: it is fat and short, and hence does not improve the query efficiency. Therefore, we propose a superior approach that partitions SE-Tree into hierarchical chunks of pre-computed surface distances, called Angular Surface Index Tree (ASI-Tree). Unlike SE-tree, ASI-Tree is a well balanced thin and tall tree. With ASI-Tree, we can continuously monitor skNN queries efficiently with low CPU and I/O overheads by both speeding up the surface shortest path computations and localizing the searches. We experimentally verify the applicability and evaluate the efficiency of the proposed methods with both real world and synthetic data sets. ASI-Tree consistently and significantly outperforms SE-Tree in all cases.