Incremental dimension reduction for efficient and accurate visual anomaly detection
This work addresses a computational bottleneck for researchers and practitioners applying deep learning-based anomaly detection to large-scale visual data, though it is incremental as it builds on existing truncated SVD methods.
The paper tackles the challenge of high-dimensional feature vectors in visual anomaly detection by introducing an incremental dimension reduction algorithm that processes data in batches, enabling efficient training on large datasets with thousands of images while maintaining close accuracy to state-of-the-art methods.
While nowadays visual anomaly detection algorithms use deep neural networks to extract salient features from images, the high dimensionality of extracted features makes it difficult to apply those algorithms to large data with 1000s of images. To address this issue, we present an incremental dimension reduction algorithm to reduce the extracted features. While our algorithm essentially computes truncated singular value decomposition of these features, other than processing all vectors at once, our algorithm groups the vectors into batches. At each batch, our algorithm updates the truncated singular values and vectors that represent all visited vectors, and reduces each batch by its own singular values and vectors so they can be stored in the memory with low overhead. After processing all batches, we re-transform these batch-wise singular vectors to the space spanned by the singular vectors of all features. We show that our algorithm can accelerate the training of state-of-the-art anomaly detection algorithm with close accuracy.