Robust object tracking for inventory monitoring systems
This project took place in winter term 2021, you CAN NOT apply to this project anymore!
Results of this project are explained in detail in the final report and presentation.
- Sponsored by: PreciBake
- Project Lead: Dr. Ricardo Acevedo Cabra
- Scientific Lead: M.Sc. Mathias Sundholm and Hamdi Belhassen
- TUM Co-Mentor: Dr. Ricardo Acevedo Cabra
- Term: Winter semester 2021
PreciBake is a company based in Munich, New York and Mumbai, developing AI solutions for the food-tech and baking industry. Our AI team is continuously working on developing and improving our machine learning algorithms for tasks such as image classification, object detection and tracking.
Computer vision based inventory monitoring systems can be used by bakeries and restaurants to track information about the inventory, such as its size and age, in real-time. This kind of system needs to rely on robust detection and tracking of items and their trajectories through space and time to work correctly.
Multi-object tracking for inventory monitoring systems is a challenging computer vision problem since frame to frame objects correspondances have to be made reliably for longer periods of time. Additionally objects might get partially or temporarily occluded by people adding or removing products from the inventory.
The goal of this project is to explore and develop deep learning based object tracking algorithms that reliably can track food moving in and out of the inventory.
Accepted students to this project should attend (unless they have proven knowledge) online workshops at the LRZ from 11.10. - 15.10.2021. More information will be provided to students accepted to this project.