The rapid evolution of deep learning and multimodal foundation models has propelled video understanding and information retrieval to the forefront of artificial intelligence. Unlike static images, videos integrate complex visual, temporal, and auditory signals, making them one of the most informative yet challenging data modalities for machine perception. As video content continues to dominate global internet traffic, developing robust methods for effectively indexing, searching, and reasoning over this high-dimensional data is of paramount importance for both academia and industry.

We are pleased to invite you to contribute to this Special Issue, which aims to bring together cutting-edge research that advances the representation, understanding, and retrieval of video and multimodal content. This Special Issue aligns with the journal’s scope by fostering interdisciplinary perspectives across computer vision, multimedia, and information retrieval, seeking to provide a comprehensive roadmap for the future of video-centric multimodal intelligence.

In this Special Issue, original research articles and reviews are welcome. Research areas may include (but are not limited to) the following:

• Video-text retrieval and cross-modal alignment;
• Large-scale video pretraining and foundation models (e.g., Video MLLMs);
• Efficient indexing and ranking algorithms for large-scale multimedia databases;
• Generative modeling for video retrieval and synthesis;
• Robust evaluation benchmarks for video understanding;
• Agentic AI and visual reasoning;
• Real-world applications in robotics, healthcare, and human–computer interaction.

I look forward to receiving your contributions.