Road Damage Detection: A Practical Case-Study of Computer Vision

The advancement in the field of artificial intelligence, specifically computer vision, has enabled researchers to approach previously unsolvable tasks such as road damage detection. In a practical sense, automated evaluation of road damage offers the ability to effectively detect hazardous road conditions, and efficiently repair said roads. Many institutions have made this task a priority focus and have published incentives for researchers to develop technologies regarding road damage detection in the form of research competitions. Specifically, this seminar will discuss the “Road Damage Detection” task from the Institute of Electrical and Electronics Engineers’ (abbrev. IEEE) Big Data 2020 Competition. In a technical sense, this seminar will focus on various computer vision techniques that have been previously applied to similar problems, and current cutting-edge technologies that present the best success in road damage detection.

Another aspect of this seminar will be a focus on a pragmatic approach to solving structured technical problems, such as the “Road Damage Detection” task from the IEEE. Application of theoretical knowledge is often overshadowed by the complexity of the overwhelming nature of previously ‘impossible’ tasks, even in an artificial setting. This seminar will offer insight into solving heterogeneous problems with a pragmatic and practical approach in a technical context. 

Link to meeting: https://meet.google.com/jbj-tdsr-aop