CDVA datasets
1.1.1.1.1.1.1.1.34m36813 AHG on Compact Descriptors for Video Analysis [Convenor]
1.1.1.1.1.1.1.1.35m36246 RAI Dataset for CDVA: Monuments of Italy [Daniele Buzzi, Angela Maria de Cristofaro, Grazia de Santis, Alberto Messina, Giacomo Perrozzi, Silvia Proscia]
RAI Teche (RAI's Archive department) started a massive campaign of selection of material from their archives, and identified of a set of about two hundred objects (i.e., monuments, buildings, statues), for each which a variable number of archive clips have been also identified. These objects, that can be collectively classified as large objects and are geographically located in all the regions of the country. There are approximately 10 objects per region. For future reference, the dataset is named “RAI Monuments of Italy”. At present approximately 50% of the dataset have been collected.
1.1.1.1.1.1.1.1.36m36765 Visual Atoms Contribution to CDVA Datasets [Stavros Paschalakis, Kevin Jouneau]
Telecom Italia provided a set of 1800 video clips, containg 180 distinct buildings, with 10 video clips of each building. The video clips all have duration of approximately 1 second. The material falls into the category “Large objects like building facades, landmarks, etc.” The same illumination conditions and camera were used throughout.
Dostları ilə paylaş: |
