Grand Challenge
Title: Multimodal Imaging for Waste Detection and Sorting
Abstract: Organic waste management is a crucial component of a circular economy, which prioritizes reducing waste through the reuse and recycling of products and materials. It is a tedious and complicated task, largely accomplished through manual labor. We introduce a novel `in-the-wild’ multimodal image dataset of 15-band NIR multi-spectral and single band thermal images of bulk food waste in an industrial setting. The dataset showcases several complex computer vision problems that are unavoidable constraints in this setting. Benchmarking against different computer vision algorithms is performed to highlight these challenges. The key issues and their place in robotic waste processing for industrial applications, and grand challenge objectives are discussed. As this is the inaugural edition of the challenge, the primary objective is not to establish a strong baseline, but to showcase and explore particular computer vision challenges that are characteristic of the waste sorting issue. The dataset provided for use is the Thermal and Spectral Trash Yield (TaSTY), which consists of roughly 900 labelled and annotated multi modal images.
Program: The challenge will be held on the 15th of October as part of the main conference program.
- 3:30pm: 5 min – Welcome and intro
- 3:35pm: 25 min – Keynote speaker: Olympia Yarger
- 4:00pm: 15 min – Matthew Vestal: Introduction to Grand Challenge and summary paper presentation
- 4:15pm: 10 min – Elane Peng: mIoG: An Evaluation Metric for Multispectral Instance Segmentation in Robotics
- 4:25pm: 5min – Concluding remarks
Organizers:
- Matthew Vestal (University of Canberra)
-
James Ireland (University of Canberra)
-
Damith Herath (University of Canberra)
-
Xing Wang (University of Canberra)
-
Ram Subramanian (University of Canberra)