Understanding human activity and thus effectively collaborating with humans is critical for some artificial intelligence systems. It is a very challenging problem which involves multiple tasks such as human action detection, person tracking, pose estimation, human-object interaction, and so on. Each of them has been independently developed into a research sub-area. Among them, however, there may exist some connections, which can be leveraged to boost the recognition. The purpose of this workshop is to bring together the research on human activity understanding, which hopefully can trigger more discussions on cross-task recognition and inspire new research ideas for human-centric activity understanding. This workshop encourages multi-task pattern recognition research, such as joint action detection and person tracking, joint event segmentation and recognition, joint pose tracking and estimation, and so on.
Workshop topics include, but are not limited to, the following:
Accepted papers will be included in the ICPR 2020 Workshop Proceedings, which will be published by Springer in the Lecture Notes in Computer Science Series (LNCS).
The submission format should follow the Springer LNCS layout.
The page limit for full and short papers are as follows:
Springer LNCS template for submission in Word and Latex can be downloaded here:
Springer's proceedings LaTeX templates are also available in Overleaf.
The submission site is available at: CMT submission site.