Learning To Anticipate Human Actions That Happen In The Next One To Five Seconds Using Self-Regularised Future Content Generation For Human-Robotic Engagement

Project Description

Our ability to anticipate the behaviour of others comes naturally to us. For example, an experienced driver can often predict the behaviour of other road users. This phenomenon is called Action Anticipation, the ability to recognise actions of others before it happens in the immediate future.

This is so natural to us but how to develop a computational approach to do the same remains a challenge. It is critical to transfer this ability to computers so that robots may be able to react quickly by anticipating human actions like humans. Robots’ ability to understand what humans might do in the immediate future is important for the development of assistive robotics in domains such as manufacturing and healthcare. The objective of this project is to investigate a novel approach to anticipate human actions, specifically one to five seconds before action happens using visual information in human-robot engagement scenarios.

Research Technical Area

  • Computer vision
  • Machine learning

Benefits to the society

The potential findings of this project are useful for future warning systems in human-robotic engagement applications such as automated driving and human-robotic assembly tasks. Furthermore, project finding can be exploited in robotic systems to plan ahead for human behaviour.

Project’s Publications

Team’s Principal Investigator

Dr Basura Fernando
Agency for Science, Technology and Research

Basura Fernando is a research scientist at the Artificial Intelligence Initiative (A*AI) of Agency for Science, Technology and Research (A*STAR) Singapore. Prior to that he was a research fellow at the Australian National University and a project leader at the Australian Centre for Robotic Vision (ACRV) where he lead the project  titiled “Understanding Human and Robot Actions and Interactions”. He obtained PhD from VISICS group of KU Leuven, Belgium in March 2015 under the supervision of Professor Tinne Tuytelaars. He is interested in Computer Vision and Machine Learning research.

 

Recent Notable Awards

  • ICCV 2017 Best Reviewer Award