Bridging Theory and Practice: Undergraduate Engagement in Computer Vision and Robotics

Ayush Vasu Gowda; Daniel Raviv; Juan David Yepes

Download Paper | Permalink

Conference: 2025 ASEE Southeast Conference
Location: Mississippi State University, Mississippi
Publication Date: March 9, 2025
Start Date: March 9, 2025
End Date: March 11, 2025
Conference Session: Professional Papers
Tagged Topic: Professional Papers
Page Count: 16
DOI: 10.18260/1-2--54206
Permanent URL: https://peer.asee.org/54206
Download Count: 33

Dr. Raviv is a Professor of Computer & Electrical Engineering and Computer Science at Florida Atlantic University. In December 2009 he was named Assistant Provost for Innovation and Entrepreneurship.

With more than 30 years of combined experience in th

visit author page

biography

Juan David Yepes Florida Atlantic University

visit author page

Assistant Professor in Teaching

visit author page

Download Paper | Permalink

Abstract

This paper presents the journey of a computer engineering undergraduate student venturing into the field of computer vision and robotics, with a focus on optical flow and its applications in detecting moving objects during translational camera motion. It highlights the technical challenges encountered, along with the strategies employed to overcome them. Beyond technical skill acquisition, the paper also emphasizes the development of interpersonal skills, such as teamwork and navigating adversity. Additionally, the paper offers educators a hands-on experiment that can be implemented in the classroom, enabling students to engage with the same concepts and gain practical experience in an accessible way. In this paper, we detail the progression of technical expertise, problem-solving abilities, and creative thinking fostered through exploration.

The student joined this project with minimal robotics knowledge and only a basic understanding of computer vision. He learned about theoretical mathematical algorithms developed prior to his involvement and was introduced to existing Python and Excel simulations. After learning the theory, the student assembled a Hiwonder JetAuto Pro Jetson Nano robot, created an artificial 3D environment, developed a Python program using OpenCV, and implemented and verified the theories and simulations. He also recorded and processed relevant videos.

As part of a team consisting of a professor, an electrical engineering PhD student, and other undergraduates, the student participated in weekly meetings where they discussed various problem-solving ideas using both divergent and convergent thinking. These meetings facilitated the exchange of ideas, yielding multiple solutions. One such solution involved integrating a mobile phone with the Hiwonder robot to capture and synchronize accelerometer and gyroscope data.

This paper traces the student's exploration, which extended beyond classroom learning into hands-on experimentation using state-of-the-art robotic systems. We suggest how to expand this success to many students interested in hands-on research. To make the concepts of optical flow accessible to students, we developed a hands-on classroom activity for classroom implementation. Students set up a smartphone or webcam camera to record videos of objects moving toward or away from it at various speeds. They then import these videos into their IDEs and apply optical flow algorithms like Lucas-Kanade or Farneback to visualize and analyze the motion patterns. By observing the optical flow vectors, students directly experience the effects of visual looming. This activity reinforces theoretical concepts, provides practical experience with image processing techniques, and fosters critical thinking as students troubleshoot and engage with advanced computer vision topics.

Citation
Format

Gowda, A. V., & Raviv, D., & Yepes, J. D. (2025, March), Bridging Theory and Practice: Undergraduate Engagement in Computer Vision and Robotics Paper presented at 2025 ASEE Southeast Conference , Mississippi State University, Mississippi. 10.18260/1-2--54206

TY - CPAPER
AB - This paper presents the journey of a computer engineering undergraduate student venturing into the field of computer vision and robotics, with a focus on optical flow and its applications in detecting moving objects during translational camera motion. It highlights the technical challenges encountered, along with the strategies employed to overcome them. Beyond technical skill acquisition, the paper also emphasizes the development of interpersonal skills, such as teamwork and navigating adversity. Additionally, the paper offers educators a hands-on experiment that can be implemented in the classroom, enabling students to engage with the same concepts and gain practical experience in an accessible way. In this paper, we detail the progression of technical expertise, problem-solving abilities, and creative thinking fostered through exploration.

The student joined this project with minimal robotics knowledge and only a basic understanding of computer vision. He learned about theoretical mathematical algorithms developed prior to his involvement and was introduced to existing Python and Excel simulations. After learning the theory, the student assembled a Hiwonder JetAuto Pro Jetson Nano robot, created an artificial 3D environment, developed a Python program using OpenCV, and implemented and verified the theories and simulations. He also recorded and processed relevant videos.

As part of a team consisting of a professor, an electrical engineering PhD student, and other undergraduates, the student participated in weekly meetings where they discussed various problem-solving ideas using both divergent and convergent thinking. These meetings facilitated the exchange of ideas, yielding multiple solutions. One such solution involved integrating a mobile phone with the Hiwonder robot to capture and synchronize accelerometer and gyroscope data.

This paper traces the student&#39;s exploration, which extended beyond classroom learning into hands-on experimentation using state-of-the-art robotic systems. We suggest how to expand this success to many students interested in hands-on research. To make the concepts of optical flow accessible to students, we developed a hands-on classroom activity for classroom implementation. Students set up a smartphone or webcam camera to record videos of objects moving toward or away from it at various speeds. They then import these videos into their IDEs and apply optical flow algorithms like Lucas-Kanade or Farneback to visualize and analyze the motion patterns. By observing the optical flow vectors, students directly experience the effects of visual looming. This activity reinforces theoretical concepts, provides practical experience with image processing techniques, and fosters critical thinking as students troubleshoot and engage with advanced computer vision topics.

AU - Ayush Vasu Gowda
AU - Daniel Raviv
AU - Juan David Yepes
CY - Mississippi State University, Mississippi
DA - 2025/03/09
PB - ASEE Conferences
TI - Bridging Theory and Practice: Undergraduate Engagement in Computer Vision and Robotics
UR - https://peer.asee.org/54206
DO - 10.18260/1-2--54206
ER -