MVTec Dataset
收藏资源简介:
MVTec数据集是一个专为评估异常检测算法而设计的大型图像数据集,包含10种对象和5种纹理,适用于测试各种异常检测技术的鲁棒性和泛化能力。该数据集在持续学习场景中被用于评估像素级异常检测的性能,特别是在新数据不断到来的情况下。数据集的创建旨在解决工业和医疗领域中不断出现的新对象和结构的异常检测问题,强调了在非静态数据分布下进行有效异常检测的重要性。
The MVTec Dataset is a large-scale image dataset specifically designed for evaluating anomaly detection algorithms, encompassing 10 object categories and 5 texture categories. It is suitable for testing the robustness and generalization capability of various anomaly detection techniques. This dataset is utilized in continual learning scenarios to assess the performance of pixel-level anomaly detection, especially when new data arrives continuously. The dataset was created to address the anomaly detection problems of novel objects and structures continuously emerging in industrial and medical fields, emphasizing the significance of effective anomaly detection under non-stationary data distributions.

- 1Unveiling the Anomalies in an Ever-Changing World: A Benchmark for Pixel-Level Anomaly Detection in Continual Learning帕多瓦大学 · 2024年



