Autors: Ismailov, A. V., Hristov, V. D. Title: YOLO Performance Comparison on Stock Images of Grocery Products Keywords: classification, grocery products, stock images, YOLOAbstract: This paper compares the classification performance of the YOLO models in the cases where the input data is limited or purely generated from stock images. For this purpose, YOLOv5n, YOLOv5s, YOLOv8n, and YOLOv8s were trained to classify stock images into 219 categories of grocery products. References - S. Wang and Z. Su, "Metamorphic Testing for Object Detection Systems, " arXiv [cs.CV], 2019.
- P. Jund, N. Abdo, A. Eitel, and W. Burgard, "The Freiburg Groceries Dataset, " vol. abs/1611.05799, 2016.
- M. Klasson, C. Zhang, and H. Kjellstrom. "A Hierarchical Grocery Store Image Dataset with Visual and Semantic Labels, " arXiv [cs.CV], 2019.
- H. Khan, Workspace for grocery store product classifier development. [Online]. Available: https://github.com/hNao17/product-classification. [Accessed Oct. 7, 2024]
- M. Klasson, C. Zhang, and H. Kjellstrom. "Using Variational Multiview Learning for Classification of Grocery Items, " Patterns, vol. 1, no 8, pp. 100143, 2020.
- A. Ismailov, V. Hristov, "Brochure segmentation methodology, " presented in the International Conference Automatics, Robotics & Artificial Intelligence ICARAI 2024, Sozopol, Bulgaria, 2024.
- G. Jocher, Ultralytics YOLOv5. 2020. [Online]. Available: https://github.com/ultralytics/yolov5. [Accessed Oct. 7, 2024]
- G. Jocher, A. Chaurasia, and J. Qiu, Ultralytics YOLOv8. 2023. [Online]. Available: https://github.com/ultralytics/ultralytics. [Accessed Oct. 7, 2024]
Issue
| National Conference with International Participation, TELECOM, 2025, Bulgaria, https://doi.org/10.1109/TELECOM63374.2024.10812185 |
|