Depth-based Computer Vision for In-vehicle Monitoring

A course for PhD students
Alma Mater Studiorum Università di Bologna
DISI – via dell’Università 50, Cesena

6, 7, 8 April 2022

Guido Borghi, PhD
Assistant Professor (RTD-A)
Dipartimento di Informatica – Scienza e Ingegneria
University of Bologna

Please, register here.

About the course
The loss of vehicle control is a common problem, due to driving distractions, stress, fatigue and bad psycho-physical conditions. Furthermore, the future arrival of (semi-) self-driving cars and the necessary transition period, characterized by the coexistence of traditional and autonomous vehicles, is going to increase the already-high interest about driver attention monitoring systems. The course will introduce Driver Monitoring principles through Computer Vision, Deep Learning techniques and depth data. Then, we  will describe learning-based methods to estimate the pose of the head and, in general, of the human body of the driver. Finally, we will see real use cases applied on modern literature datasets.

Assessment modalities
Students requiring an exam will be asked to produce a written essay on a relevant topic, agreed with the instructor, about possible intersections between their research interests and the topics/open issues presented in this course. In case the produced paper would be submitted for consideration to a scientific journal, students are allowed to write it in groups. A Driver Monitoring prototype implementation may be also considered by the instructor as an assessment modality.

Teaching materials
Lecture notes and slides will be provided by the instructor, along with papers and a list of bibliographical references and additional material. All the course material is in English.

Material is provided through .zip encrypted files (AES-256), I suggest using 7zip (or similar programs) to extract them.


    • The importance of Driver Monitoring Systems in the Self-Driving Cars era
    • Depth data acquisition inside vehicles
    • Comparison of depth devices
    • Deep-learning for Head and Body Pose Estimation task
    • Deep-learning for Facial Landmark Detection
    • Deep-learning for Head/Face Detection
    • Analysis of available datasets in the literature
    • Real use-cases and ADAS systems

Lectures will provided both online (through Microsoft teams) and with frontal lessons at DISI – via dell’Università 50, Cesena.

Students that will partecipate in frontal lessons must have the Green Pass certificate.

  • Wednesday 06/04/2022
    • 09.00 – 13.00
    • Room: 3.10
    • Microsoft Teams link
  • Thursday 07/04/2022
    • 10.00 – 13.00
    • Room: 2.13
    • Microsoft Teams link
  • Friday 08/04/2022
    • 10.00 – 13.00
    • Room: 2.13
    • Microsoft Teams link