Human and machine vision

Teachers
Gianluca Campana
Dept. of General Psychology, University of Padua
gianluca.campana[at]unipd.it
M-PSI/01


Lamberto Ballan
Dept. of Mathematics, University of Padua
lamberto.ballan[at]unipd.it
INF/01

Aim
This course will show the basic mechanisms and processing of human vision, with particular emphasis to the increasingly complexity of receptive fields along the visual hierarchy, and the way such mechanisms have been implemented in computers, mostly by the definition of deep neural networks. This topic is one of the most fruitful examples of integration between cognitive neuroscience and computer science. Computer vision techniques are currently revolutionizing many scientific and industrial sectors, such as automotive and mobility, healthcare, security, robotics and automation.
This course will describe the notion of user experience and its dimensions, will consider the most common methods in user-based design and evaluation, will  skecth some of the social aspects of computing and will do so in connection with the students' specific projects.

Syllabus
- The concept of receptive field (prof Campana)
- Parvo and magno pathways (prof Campana)
- From the retina to the striate cortex (prof Campana)
- Types of receptive field in striate cortex: detectors of bars or detectors of spatial frequencies? (prof Campana)
- Receptive fields in extrastriate cortices (prof Campana)
- Machine perception and computer vision (prof Ballan)
- MFrom image processing and early vision to high-level visual recognition (prof Ballan)
- MA brief introduction to convolutional neural networks and their applications (ranging from security and surveillance, smart mobility, robotics, self-driving cars, etc.) (prof Ballan)

Course requirements
None

Introductory readings
https://www.cns.nyu.edu/~david/courses/perception/lecturenotes/V1/lgn-V1.html

https://www.youtube.com/watch?v=JIBvfxg2iJ0&list=PLd73TvRLcri_KlqcCSYCAnv4pi-ZXBrBy

Examination modality
Students are asked to study and work in small interdisciplinary groups (2-3 students) on a scientific article relevant to the content of the course. They will present the paper and discuss it with the instructors and their peers in the last lecture of the course.

Course material, enrollment and last minute notifications
Made available by the teacher at this Moodle address

Schedule
19 April 2021, 9:30-12:00
20 April 2021, 9:30-12:00
22 April 2021, 9:30-12:00
23 April 2021, 9:30-12:00
29 April 2021, 9:30-12:00

Location
Room 1BC50, Dept of Mathematics; Zoom link in case of quarantined students will be in the Moodle page of the course.

<< Courses in 2020-2021