用流模型检测图像场景图中的异常关系，速度快且更稳定。
BUSSARD: Normalizing Flows for Bijective Universal Scene-Specific Anomalous Relationship Detection
- 用归一化流学习物体-关系-物体三元组的双射变换
- 在SARD数据集上比当前最佳模型高10%的AUROC，快5倍
- 对同义词不敏感，性能波动比基线低17.5%
我们提出基于归一化流的通用场景特定异常关系检测方法BUSSARD，用于从图像生成的场景图中检测异常关系。该方法采用多模态策略，通过语言模型嵌入场景图中的物体与关系标记，利用真实世界语义知识。使用归一化流模型学习双射变换，将物体-关系-物体三元组映射到简单基础分布（通常为高斯分布），通过似然估计实现异常检测。我们在包含办公室和餐厅场景的SARD数据集上进行评估，结果表明，本方法相比当前最优模型在AUROC上提升约10%，同时速度提高五倍。消融实验显示，本方法在同义词使用方面具有更强鲁棒性和通用性，性能波动远低于基线（仅17.5%偏差）。该工作展示了基于学习的方法在场景图关系异常检测中的巨大潜力。代码已开源：https://github.com/mschween/BUSSARD。
原文摘要 · Abstract (English)
We propose Bijective Universal Scene-Specific Anomalous Relationship Detection (BUSSARD), a normalizing flow-based model for detecting anomalous relations in scene graphs, generated from images. Our work follows a multimodal approach, embedding object and relationship tokens from scene graphs with a language model to leverage semantic knowledge from the real world. A normalizing flow model is used to learn bijective transformations that map object-relation-object triplets from scene graphs to a simple base distribution (typically Gaussian), allowing anomaly detection through likelihood estimation. We evaluate our approach on the SARD dataset containing office and dining room scenes. Our method achieves around 10% better AUROC results compared to the current state-of-the-art model, while simultaneously being five times faster. Through ablation studies, we demonstrate superior robustness and universality, particularly regarding the use of synonyms, with our model maintaining stable performance while the baseline shows 17.5% deviation. This work demonstrates the strong potential of learning-based methods for relationship anomaly detection in scene graphs. Our code is available at https://github.com/mschween/BUSSARD .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。