Our lab name is "Multimodal Analysis, Retrieval, and Search (MARS)." MARS focuses on advancing technologies that enable machines to understand, index, and retrieve information from diverse data sources. Our research integrates methods from information retrieval, multimedia indexing, data mining, and computer vision to address the growing challenges of managing large-scale, heterogeneous, and multimodal datasets.

We explore innovative solutions for effective and efficient retrieval across multiple data modalities, such as text, images, audio, and video. By combining deep learning techniques with traditional retrieval models, our work aims to bridge the semantic gap between different data types and improve user-centric search experiences. Our current projects span from cross-modal retrieval and recommender systems to large-scale data mining and visual understanding tasks.

Through interdisciplinary collaboration and applied research, our lab strives to develop scalable and robust algorithms that push the boundaries of multimodal information processing, contributing to both academic advancements and real-world applications.