| dc.rights.license | CC-BY-NC-ND | |
| dc.contributor.advisor | Salah, Albert | |
| dc.contributor.author | Abdalla Mohamed Salama Sayed Moustafa, Abdalla | |
| dc.date.accessioned | 2023-09-06T10:07:47Z | |
| dc.date.available | 2023-09-06T10:07:47Z | |
| dc.date.issued | 2023 | |
| dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/45031 | |
| dc.description.sponsorship | Utrecht University | |
| dc.language.iso | EN | |
| dc.subject | This thesis is done with collaboration with Orbisk to tackle food waste image processing. More specifically, the thesis proposes a multi-task framework that can handle segmentation, classification and regression tasks concurrently. The framework is trained and validated on Orbisk's own dataset, which surpasses existing benchmark datasets on food images. The results show a strong performance on all tasks, specially the instance-segmentation task. | |
| dc.title | FoodWasteAI: A Multi-Task Transformer Framework For Food Waste Image Processing | |
| dc.type.content | Master Thesis | |
| dc.rights.accessrights | Open Access | |
| dc.subject.keywords | Food waste; Computer vision; Deep learning; Instance segmentation; Vision transformers; Multi-task learning | |
| dc.subject.courseuu | Artificial Intelligence | |
| dc.thesis.id | 23550 | |
