SALAD -- Semantics-Aware Logical Anomaly Detection
Not on the map. Why?
Abstract (arXiv)
Recent surface anomaly detection methods excel at identifying structural anomalies, such as dents and scratches, but struggle with logical anomalies, such as irregular or missing object components. The best-performing logical anomaly detection approaches rely on aggregated pretrained features or handcrafted descriptors (most often derived from composition maps), which discard spatial and semantic information, leading to suboptimal performance. We propose SALAD, a semantics-aware discriminative logical anomaly detection method that incorporates a newly proposed composition branch to explicitly model the distribution of object composition maps, consequently learning important semantic relationships. Additionally, we introduce a novel procedure for extracting composition maps that requires no hand-made labels or category-specific information, in contrast to previous methods. By effectively modelling the composition map distribution, SALAD significantly improves upon state-of-the-art methods on the standard benchmark for logical anomaly detection, MVTec LOCO, achieving an impressive image-level AUROC of 96.1%. Code: https://github.com/MaticFuc/SALAD
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 33 of its 47 references that are in the map. First posted on arXiv 2 Sep 2025.
Builds on
33 of 47 references.
References · 33
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
- 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
- 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
- 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
- 2017Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentationscited by 614
- 2021DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detectioncited by 510
- 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 419
- 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 397
- 2023Segment Anything in High Qualitycited by 373
- 2021CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flowscited by 320
- 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
21 more
- 2021FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flowscited by 225
- 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 157
- 2020Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detectioncited by 115
- 2022Asymmetric Student-Teacher Networks for Industrial Anomaly Detectioncited by 115
- 2021Mixed supervision for surface-defect detection: from weakly to fully supervised learningcited by 101
- 2021Inpainting Transformer for Anomaly Detectioncited by 91
- 2022DSR -- A dual subspace re-projection network for surface anomaly detectioncited by 89
- 2022Omni-frequency Channel-selection Representations for Unsupervised Anomaly Detectioncited by 79
- 2022MemSeg: A semi-supervised method for image surface defect detection using differences and commonalitiescited by 68
- 2023TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detectioncited by 43
- 2023Focus the Discrepancy: Intra- and Inter-Correlation Learning for Image Anomaly Detectioncited by 32
- 2025LogicAD: Explainable Anomaly Detection via VLM-based Text Feature Extractioncited by 24
- 2023Component-aware anomaly detection framework for adjustable and logical industrial visual inspectioncited by 22
- 2023Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detectioncited by 19
- 2023Contextual Affinity Distillation for Image Anomaly Detectioncited by 18
- 2023Generating and Reweighting Dense Contrastive Patterns for Unsupervised Anomaly Detectioncited by 14
- 2024SAM-LAD: Segment Anything Model Meets Zero-Shot Logic Anomaly Detectioncited by 14
- 2024CSAD: Unsupervised Component Segmentation for Logical Anomaly Detectioncited by 11
- 2023SLSG: Industrial Image Anomaly Detection by Learning Better Feature Embeddings and One-Class Classificationcited by 10
- 2024PUAD: Frustratingly Simple Method for Robust Anomaly Detectioncited by 7
- 2023Set Features for Anomaly Detectioncited by 3
Built on by
Citing papers per year: 2026: 3.
3 papers.