Multimodal cues for Driver's Distraction

Adirondack StreamIn the Multimodal Signal Processing lab at UT Dallas, I am working on modelling the driver's distraction which will help in designing active in-vehicle safety systems. By monitoring the driver's action his state of mind can be understood. This information can help warn about any impeding danger and save from potential accidents.

We are currently looking at the headpose of the driver to get information about the visual attention. Some of the questions we are looking to answer is, how much disparity is there between the driver's exact gaze direction and his headpose. How much does the head pose varies when the driver is looking in the certain direction. How well can the headpose be used to track the attention of the driver.

Publications

  • Sumit Jha and Carlos Busso, "Analysis of head pose as an indicator of drivers' visual attention," in 7th Biennial Workshop on DSP for In-Vehicle Systems and Safety, Berkeley, CA, USA, October 2015.
  • Sumit Jha and Carlos Busso, "Analyzing the Relationship Between Head Pose and Gaze to Model Driver Visual Attention," in IEEE Conference on Intelligent Transportation Systems (ITSC 2016), Rio de Janeiro, Brazil, November 2016.