Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
ReplayCAD: Generative Diffusion Replay for Continual Anomaly Detection
Authors: Lei Hu, Zhiyong Gan, Ling Deng, Jinglin Liang, Lingyu Liang, Shuangping Huang, Tianshui Chen
IJCAI 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our method achieves state-of-the-art performance in both classification and segmentation, with notable improvements in segmentation: 11.5% on Vis A and 8.1% on MVTec. Our source code is available at https://github.com/HULEI7/Replay CAD. Extensive experiments demonstrate that our method significantly outperforms SOTA approaches, particularly in segmentation performance, achieving improvements of 11.5% and 8.1% on Vis A and MVTec, respectively. Table 1: Comparison with state-of-the-art methods. |
| Researcher Affiliation | Collaboration | 1South China University of Technology 2China United Network Communications Corporation Limited Guangdong Branch 3Pazhou Laboratory 4Guangdong University of Technology |
| Pseudocode | No | The paper describes the methodology using textual descriptions and architectural diagrams (Figures 4 and 5) but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | Our source code is available at https://github.com/HULEI7/Replay CAD. |
| Open Datasets | Yes | Datasets. We conducted experiments on two commonly used industrial anomaly detection datasets: Vis A [Zou et al., 2022] and MVTec [Bergmann et al., 2019]. |
| Dataset Splits | Yes | Vis A comprises 12 sample classes, with 8659 training samples (all normal) and 2162 test samples (962 normal, 1200 anomalous). MVTec includes 15 sample classes, with 3629 training samples (all normal) and 1725 test samples (467 normal, 1258 anomalous). When learning the i-th class, we can only access the training data Xi and cannot access data from previous classes {X1, X2, , Xi 1}, during testing, we need to perform testing on {X1, X2, , Xi}. |
| Hardware Specification | No | The paper does not explicitly describe the hardware used for running its experiments. It mentions using pre-trained Stable Diffusion and LDM weights but no details about the computational resources. |
| Software Dependencies | No | The paper mentions using 'pre-trained Stable Diffusion 1.5 weights' and 'pre-trained LDM weights', but it does not specify other ancillary software dependencies with version numbers (e.g., Python, PyTorch, CUDA versions). |
| Experiment Setup | Yes | We set the number of tokens K for the semantic embedding v to 20. For Vi SA, we use pre-trained Stable Diffusion 1.5 weights with a resolution of (512, 512), where the dimension C of the semantic and spatial features is 768, and the MLP layer size is (128, 196). For MVTec, we use pre-trained LDM weights with a resolution of (256, 256), where the dimension C of the semantic and spatial features is 1280, and the MLP layer size is (128, 200). We generate 800 samples for each class for training the anomaly detection model. We use Invad [Zhang et al., 2024b] as the baseline model for anomaly detection, keeping all settings consistent with the original paper. |