Odd-One-Out:通过与邻域对象对比进行异常检测
Odd-One-Out: Anomaly Detection by Comparing with Neighbors
AI总结:
本研究提出一种基于场景内对象对比的新型异常检测任务,通过多视图构建3D对象中心模型并结合部件感知表示,经跨实例对比检测场景特有异常,同时发布两个新基准并完成全面验证。
摘要:
本文提出了一种新的异常检测(AD)问题,旨在通过将场景内的对象与其他存在的对象进行对比,识别其中“外观怪异”的对象。与采用固定异常判定标准的传统AD基准不同,本任务通过推断常规对象的参考组,检测每个场景特有的异常。为解决遮挡问题,我们将每个场景的多视图作为输入,从2D视图为每个实例构建以3D对象为中心的模型,并通过几何一致的部件感知表示增强这些模型,随后通过跨实例对比检测异常对象。我们还推出了两个新基准ToysAD-8K和PartsAD-15K,作为该任务未来研究的测试平台,并在这些基准上对所提方法进行了定量和定性的全面分析。
英文摘要:
This paper introduces a novel anomaly detection (AD) problem aimed at identifying `odd-looking' objects within a scene by comparing them to other objects present. Unlike traditional AD benchmarks with fixed anomaly criteria, our task detects anomalies specific to each scene by inferring a reference group of regular objects. To address occlusions, we use multiple views of each scene as input, construct 3D object-centric models for each instance from 2D views, enhancing these models with geometrically consistent part-aware representations. Anomalous objects are then detected through cross-instance comparison. We also introduce two new benchmarks, ToysAD-8K and PartsAD-15K as testbeds for future research in this task. We provide a comprehensive analysis of our method quantitatively and qualitatively on these benchmarks.
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。