Simplified Distracted Driving Detection with Facial Keypoints
Faculty Mentor
Dr. Rami Haddad
Location
Poster 207
Session Format
Poster Presentation
Academic Unit
Department of Electrical and Computer Engineering
Background
In this work, the primary goal is human isolation and pose classification. Based on this approach, keypoint detection was chosen as the CV model type. OpenPose is a popular pretrained keypoint model with an extensive GitHub and API compatibility with multiple programming languages [2]. It utilizes heatmaps to isolate body keypoints and part affinity fields to connect keypoints of the same person together. When imported and given an image, OpenPose returns a three-dimensional array that contains the x-y coordinates of each point for every person.
Keywords
Allen E. Paulson College of Engineering and Computing Student Research Symposium, Computer Vision, CV
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Presentation Type and Release Option
Presentation (File Not Available for Download)
Start Date
2022 12:00 AM
January 2022
Simplified Distracted Driving Detection with Facial Keypoints
Poster 207
According to the US National Highway Traffic Safety Administration, distracted driving was the primary cause of 3,142 fatalities in 2019. Distracted driving can be initiated by both objects and people inside the car, the current surroundings, and other unpredictable events. Therefore, it becomes necessary to develop a proactive approach to detect when a driver is not focusing on sensible objects and vehicles on the road. For this detection system to be feasible, it must be intuitive and non-invasive. Computer vision (CV), a subset of deep learning and artificial intelligence (AI), provides methods for computer systems to mimic humans in perceiving data purely from standard digital imaging.
