基于物体识别子环境，实现零样本环境理解
OBSER: Object-Based Sub-Environment Recognition for Zero-Shot Environmental Inference
- 用贝叶斯框架分析子环境与物体的关系
- 在链式检索任务中优于传统场景方法
- 适用于开放世界和真实感环境的零样本识别
我们提出物体基子环境识别（OBSER）框架，一种新型贝叶斯方法，用于推断子环境与其组成物体之间的三种基本关系。该框架利用度量学习和自监督学习模型，在潜在空间中估计子环境的物体分布，以计算这些关系。通过引入($ε,δ$)统计可分性（EDS）函数，我们在理论和实验上验证了该框架的有效性，该函数衡量表示对齐程度。OBSER框架在开放世界和逼真环境中均能可靠进行推理，在链式检索任务中表现优于基于场景的方法。该框架支持零样本环境识别，实现自主环境理解。
原文摘要 · Abstract (English)
We present the Object-Based Sub-Environment Recognition (OBSER) framework, a novel Bayesian framework that infers three fundamental relationships between sub-environments and their constituent objects. In the OBSER framework, metric and self-supervised learning models estimate the object distributions of sub-environments on the latent space to compute these measures. Both theoretically and empirically, we validate the proposed framework by introducing the ($ε,δ$) statistically separable (EDS) function which indicates the alignment of the representation. Our framework reliably performs inference in open-world and photorealistic environments and outperforms scene-based methods in chained retrieval tasks. The OBSER framework enables zero-shot recognition of environments to achieve autonomous environment understanding.
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。