ABSTRACT
Computer vision, a growing field in artificial intelligence, is crucial
in sectors such as healthcare and security. This article describes the
application of active teaching methodology in a short course on
computer vision, involving 26 Software Engineering students. The
course covered the techniques of Cascade Classifier and Histogram
of Oriented Gradients (HOG), integrating theory and practice over
20 hours of activities. The effectiveness of the course was evaluated
through quantitative and qualitative analyses. Quantitative
results showed a significant improvement in students’ performance,
while qualitative analysis indicated an increase in practical skills
and understanding of the techniques, despite challenges related to
theoretical workload and activity support. The practical approach
was praised, with suggestions to improve the teaching materials
and support during activities.