迈向增量统一多模态异常检测：从信息瓶颈视角增强多模态去噪
Towards an Incremental Unified Multimodal Anomaly Detection: Augmenting Multimodal Denoising From an Information Bottleneck Perspective
对做工业/3D 异常检测的人，这篇给出了一个可复用的思路：用信息瓶颈融合模块在融合后特征上显式过滤冗余维度，并用 Mamba 解码器解耦物体间特征耦合，从而在增量统一框架下抑制伪特征导致的灾难性遗忘。
原文依据 we introduce a novel denoising framework called IB-IUMAD, which exploits the complementary benefits of the Mamba decoder and information bottleneck fusion module
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论增量统一多模态异常检测的目标是赋予单一模型系统性地检测所有类别异常的能力，并支持增量学习以容纳新出现的物体/类别。这一追求的核心在于解决灾难性遗忘困境，即在获取新知识的同时保留先前学到的知识。尽管已有一些工作试图解决这一困境，但仍存在一个关键疏漏：忽视了虚假特征与冗余特征对灾难性遗忘的潜在影响。本文深入探究了增量统一框架中虚假特征与冗余特征对该困境的负面影响，并揭示出在相似条件下，由单模态架构朴素聚合而成的多模态框架更容易发生遗忘。为解决该问题，我们提出了一种名为 IB-IUMAD 的新型去噪框架，它利用 Mamba 解码器与信息瓶颈融合模块的互补优势：前者致力于解耦物体间的特征耦合，防止物体之间的虚假特征干扰；后者用于从融合特征中滤除冗余特征，从而显式地保留判别性信息。在 MVTec 3D-AD 与 Eyecandies 数据集上的一系列理论分析与实验证明了 IB-IUMAD 的有效性与有竞争力的性能。代码链接。
英文原文摘要 1466 字符
The quest for incremental unified multimodal anomaly detection seeks to empower a single model with the ability to systematically detect anomalies across all categories and support incremental learning to accommodate emerging objects/categories. Central to this pursuit is resolving the catastrophic forgetting dilemma, which involves acquiring new knowledge while preserving prior learned knowledge. Despite some efforts to address this dilemma, a key oversight persists: ignoring the potential impact of spurious and redundant features on catastrophic forgetting. In this paper, we delve into the negative effect of spurious and redundant features on this dilemma in incremental unified frameworks, and reveal that under similar conditions, the multimodal framework developed by naive aggregation of unimodal architectures is more prone to forgetting. To address this issue, we introduce a novel denoising framework called IB-IUMAD, which exploits the complementary benefits of the Mamba decoder and information bottleneck fusion module: the former dedicated to disentangle inter-object feature coupling, preventing spurious feature interference between objects; the latter serves to filter out redundant features from the fused features, thus explicitly preserving discriminative information. A series of theoretical analyses and experiments on MVTec 3D-AD and Eyecandies datasets demonstrates the effectiveness and competitive performance of IB-IUMAD. Code Link.