[19p-221B-1~10] 【CS.2】 Code-sharing Session of 3.3 & 4.4

▲ [19p-221B-6] [INVITED] Augmenting Visual Cognitive Interactions:
from Wearable First-person View to Ubiquitous Third-person Multi-views

〇Kenji Mase¹ (1.Nagoya Univ.)

Keywords:visual interaction, first person video processing, multi-view video recoomendation

This talk covers visual cognition and act research issues with several topics related to first-person view analysis by wearable sensing and the summarization of ubiquitously taken multi-view videos. The way of gazing and watching things contains information of personal interests in the view, which is a reflection of viewer’s knowledge and skills about the things. Such knowledge and skill will be computationally formalized to be used in wisdom computing environment. The technical challenges are in (i) locating such parts-of-interest in the view over time and (ii) finding common interest against things among people and (iii) interpreting such behavior for further utilizations.