食物识别中事后未知类别检测的基准测试
Benchmarking Post-Hoc Unknown-Category Detection in Food Recognition
AI总结：
本研究针对食物识别模型易将未知类别样本误判为分布内类别的问题，对多种事后OOD检测方法开展细粒度食物识别实证分析，发现ViM整体最优，验证了ID准确率与OOD检测性能正相关，且Transformer架构优于卷积模型。
摘要：
食物识别模型通常难以区分已见和未见样本，常将未见类别的样本错误分配一个分布内（ID）标签。这种错误分类在实际应用部署中带来了重大挑战，尤其是在自动膳食评估系统中，错误标签会导致整个系统出现级联错误。理想情况下，这类模型在遇到未知样本时应提示用户，以便采取纠正措施。由于此前尚无研究探索真实场景下的食物识别，本研究针对细粒度食物识别的多种事后分布外（OOD）检测方法开展了实证分析。研究结果表明，虚拟 logit 匹配（ViM）整体表现最佳，这可能得益于其结合了 logits 和特征空间表示。此外，本研究印证了OOD领域的已有观点，即ID准确率更高的模型在所有被评估的OOD检测方法中表现更优。而且，在多种检测方法下，基于Transformer的架构在检测OOD样本方面始终优于基于卷积的模型。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
