The context of a collective activity is characterized by all information produced during the activity. The comprehension of this context by each actor is essential for the success...
Recognizing multiple interleaved activities in a video requires implicitly partitioning the detections for each activity. Furthermore, constraints between activities are important ...
For many tasks in computer vision, it is very important to produce the groundtruth data. At present, this is mostly done manually. Manual data labeling is labor-intensive and prone...
This paper proposes a two-level theoretical framework for the study of CMC illustrating the principles of mediation and contextual analysis. Based on log files and interviews from...
: We have developed simple taxonomy of visual display-based activity in office spaces using field data. Three key user values that displays provide are identified and investigated ...