Workshop "From Pixels to Semantics - Semantic Analysis meets Visual Analysis" (PIXSEM)

Social Web Communities such as Flickr, Youtube and Facebook have collected a huge amount of valuable multimedia content. How this data can be successfully exploited has been subject of research for members from both communities - computer vision as well as knowledge mining. The aim of this workshop is to bring together outstanding research results that show how social multimedia data can be used to bridge the gap between semantic analysis and visual analysis to exploit synergies and to enable higher quality results as well as better efficiency.


  • Multimodal social media analysis and data mining
  • Multimodal learning of visual and semantic data
  • Synergies in multimodal semantic analysis
  • Multimedia event detection and event understanding
  • Analysis of social media metadata meaning and intention (semantics and pragmatics)
  • Learning from noisy data
  • Making use of context information for visual analysis
  • Semantic multimedia and visual analysis
  • Knowledge assisted multimedia analysis and data mining
  • Multimedia enriched semantic analysis
  • Content-based multimedia recommender systems
  • Named Entity Recognition and Entity Mapping for multimedia documents
  • Content-based video summarization
  • Bridging multimedia and knowledge domains
  • Semantic-driven multimedia applications

Important dates

  • Submission deadline: May 14, 2014
  • Notification of acceptance: June 6, 2014
  • Camera-ready due: June 13, 2014

More details can be found on the workshop website: