用真实产线数据重新评估工业异常检测，揭示学术方法在实际中的短板
Beyond Academic Benchmarks: Critical Analysis and Best Practices for Visual Industrial Anomaly Detection
- 基于真实生产数据建立新基准，突破实验室数据局限
- 多任务对比显示主流方法在真实场景中性能下降超30%
- 提出产学衔接的改进方向，适合工业界落地研究者参考
视觉异常检测对制造业自动化质检至关重要。当前计算机视觉领域快速发展，但多数研究依赖受控实验环境下的合成缺陷数据，难以反映真实产线复杂性。新方法常在实际部署中表现不佳，出现性能显著下降或计算资源需求过高问题。本文提出三项关键贡献：(1) 强调真实数据重要性，基于实际生产数据建立新基准；(2) 采用对实际应用有价值的指标，对现有先进方法进行跨任务公平比较；(3) 系统分析该领域进展，讨论关键挑战与弥合学术与工业差距的新视角。代码已公开于 https://github.com/abc-125/viad-benchmark。
原文摘要 · Abstract (English)
Anomaly detection (AD) is essential for automating visual inspection in manufacturing. This field of computer vision is rapidly evolving, with increasing attention towards real-world applications. Meanwhile, popular datasets are typically produced in controlled lab environments with artificially created defects, unable to capture the diversity of real production conditions. New methods often fail in production settings, showing significant performance degradation or requiring impractical computational resources. This disconnect between academic results and industrial viability threatens to misdirect visual anomaly detection research. This paper makes three key contributions: (1) we demonstrate the importance of real-world datasets and establish benchmarks using actual production data, (2) we provide a fair comparison of existing SOTA methods across diverse tasks by utilizing metrics that are valuable for practical applications, and (3) we present a comprehensive analysis of recent advancements in this field by discussing important challenges and new perspectives for bridging the academia-industry gap. The code is publicly available at https://github.com/abc-125/viad-benchmark
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。