Author: Esra Acar
Last week, we presented our paper on video affective content analysis titled as Fusion of Learned Multi-Modal Representations and Dense Trajectories for Emotional Analysis in Videos at the 13th International Workshop on Content-Based Multimedia Indexing (CBMI), Prague, Czech Republic. The aim of the workshop is to gather researchers from different communities involved in all aspects of content-based multimedia indexing, retrieval, browsing and presentation. Although there were papers about video analysis (including ours), the focus of the workshop was more about image analysis and retrieval this year. Another highlight of the workshop was that there were many papers discussing deep learning architectures for image and video analysis (including ours). More detailed information about the program can be found here.
There were also two invited speakers. The first speaker was Benjamin Bustos from University of Chile. He talked about 3D object retrieval. On the final day of the workshop, Jiri Matas from the Czech Technical University gave a keynote on diverse image retrieval. The demo that shows 3D visualization using diversified 2D images at the end of the talk was quite impressive! The slides of both talks are also provided in the CBMI web site.
Here is our presentation at CBMI:
Leave a Reply