This is a recurring event: View all events in the series “Data Bites”
Computer Vision is one of the most exciting AI technologies that, thanks to the evolution of Deep Learning, is quickly becoming one of the most powerful technologies within AI.
Due to its ability to perform many (specific) tasks at human-level accuracy, Computer Vision solutions are applied in many different contexts; retail is no exception.
In light of the increasing interest in such a technology, in this talk, we will focus on the applications of Computer Vision in retail, describing the central idea around the utilised method and highlighting the limitations of the current method.
In particular, the goal of this talk is threefold:
- Describe the value of computer vision and the different tasks it can accomplish (e.g. object detection and recognition, pose estimation and action recognition).
- Illustrate some possible applications in retail (e.g. empty shelf detection, customer behaviour analysis) underlying the scenarios where the actual solutions are not accurate enough.
- Propose some research directions aiming to solve the limitations underlined previously.
About the Speakers:
This event will have the following two speakers.
Paul Klinger
After a PhD in theoretical physics he now works on applying modern computer vision models to extract useful data from the messy real-world environment inside retail stores.
Vincenzo Crescimanna
He enjoys researching generative models and trying to understand better what the Neural Networks are learning in their hidden layers.
Attendance at City events is subject to our terms and conditions.