A5058689010- Human Mobility and Location-Based Analysis
- Video Analysis and Summarization
- Multimodal Machine Learning Applications
- Advanced Graph Neural Networks
- Graph Theory and Algorithms
- Traffic Prediction and Management Techniques
- Topic Modeling
- Geographic Information Systems Studies
- Neural Networks and Applications
- Advanced Image and Video Retrieval Techniques
Baidu (China)
2024
Text-video retrieval is a challenging task that aims to identify relevant videos given textual queries. Compared conventional retrieval, the main obstacle for text-video semantic gap between nature of queries and visual richness video content. Previous works primarily focus on aligning query by finely aggregating word-frame matching signals. Inspired human cognitive process modularly judging relevance text video, judgment needs high-order signal due consecutive complex contents. In this...
Heterogeneous graph learning aims to capture complex relationships and diverse relational semantics among entities in a heterogeneous obtain meaningful representations for nodes edges. Recent advancements neural networks (HGNNs) have achieved state-of-the-art performance by considering relation heterogeneity using specialized message functions aggregation rules. However, existing frameworks limitations generalizing across datasets. Most of these follow the "pre-train" "fine-tune" paradigm on...
Spatio-temporal prediction aims to forecast and gain insights into the ever-changing dynamics of urban environments across both time space. Its purpose is anticipate future patterns, trends, events in diverse facets life, including transportation, population movement, crime rates. Although numerous efforts have been dedicated developing neural network techniques for accurate predictions on spatio-temporal data, it important note that many these methods heavily depend having sufficient...