Research

Share         

The Rapid-Rich Object Search (ROSE) Lab was conceived in view of the proliferation of mobile internet devices and the growing need to expand search beyond text’s limited capability in describing real-world objects. The ROSE Lab was set up jointly between the Nanyang Technological University (NTU), Singapore and Peking University (PKU), China, to draw upon the respective university’s strength in media, computer vision, and cloud computing technologies, and to advance these areas further.

The lab’s research focuses on visual object search and machine learning, and the research topics include:​

​​​​
Large-scale Structured Object Database and Analytics

  • Media object database design
  • Semi-supervised and unsupervised data collection
  • Database indexing and mai​ntenance
  • Database optimization for search
 

Scalable Mobile Object Search with Contextual Mobility

  • Visual object representation and understanding
  • Fast object retrieval and recognition
  • Context-aware mobile visual search
  • Face recognition & retrieval
  • Object tracking
  • Action recognition
  • Deep Learning​
 
 
 
Media Cloud Platform

  • Cloud platform optimization for search and media services
  • GPU architecture design and optimization for machine learning
     

 

 
Media Processing

  • Video analytics
  • Image and video coding
  • Image quality measures​

​​​​