Benchmarking Post-Hoc Unknown-Category Detection in Food Recognition
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910890251517952 |
|---|---|
| author | Rahman, Lubnaa Abdur Papathanail, Ioannis Brigato, Lorenzo Mougiakakou, Stavroula |
| author_facet | Rahman, Lubnaa Abdur Papathanail, Ioannis Brigato, Lorenzo Mougiakakou, Stavroula |
| contents | Food recognition models often struggle to distinguish between seen and unseen samples, frequently misclassifying samples from unseen categories by assigning them an in-distribution (ID) label. This misclassification presents significant challenges when deploying these models in real-world applications, particularly within automatic dietary assessment systems, where incorrect labels can lead to cascading errors throughout the system. Ideally, such models should prompt the user when an unknown sample is encountered, allowing for corrective action. Given no prior research exploring food recognition in real-world settings, in this work we conduct an empirical analysis of various post-hoc out-of-distribution (OOD) detection methods for fine-grained food recognition. Our findings indicate that virtual logit matching (ViM) performed the best overall, likely due to its combination of logits and feature-space representations. Additionally, our work reinforces prior notions in the OOD domain, noting that models with higher ID accuracy performed better across the evaluated OOD detection methods. Furthermore, transformer-based architectures consistently outperformed convolution-based models in detecting OOD samples across various methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_18548 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Benchmarking Post-Hoc Unknown-Category Detection in Food Recognition Rahman, Lubnaa Abdur Papathanail, Ioannis Brigato, Lorenzo Mougiakakou, Stavroula Computer Vision and Pattern Recognition Food recognition models often struggle to distinguish between seen and unseen samples, frequently misclassifying samples from unseen categories by assigning them an in-distribution (ID) label. This misclassification presents significant challenges when deploying these models in real-world applications, particularly within automatic dietary assessment systems, where incorrect labels can lead to cascading errors throughout the system. Ideally, such models should prompt the user when an unknown sample is encountered, allowing for corrective action. Given no prior research exploring food recognition in real-world settings, in this work we conduct an empirical analysis of various post-hoc out-of-distribution (OOD) detection methods for fine-grained food recognition. Our findings indicate that virtual logit matching (ViM) performed the best overall, likely due to its combination of logits and feature-space representations. Additionally, our work reinforces prior notions in the OOD domain, noting that models with higher ID accuracy performed better across the evaluated OOD detection methods. Furthermore, transformer-based architectures consistently outperformed convolution-based models in detecting OOD samples across various methods. |
| title | Benchmarking Post-Hoc Unknown-Category Detection in Food Recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.18548 |