REVIEW 7 cited by
SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentation
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Visual anomaly detection is commonly used in industrial quality inspection. In this paper, we present a new dataset as well as a new self-supervised learning method for ImageNet pre-training to improve anomaly detection and segmentation in 1-class and 2-class 5/10/high-shot training setups. We release the Visual Anomaly (VisA) Dataset consisting of 10,821 high-resolution color images (9,621 normal and 1,200 anomalous samples) covering 12 objects in 3 domains, making it the largest industrial anomaly detection dataset to date. Both image and pixel-level labels are provided. We also propose a new self-supervised framework - SPot-the-difference (SPD) - which can regularize contrastive self-supervised pre-training, such as SimSiam, MoCo and SimCLR, to be more suitable for anomaly detection tasks. Our experiments on VisA and MVTec-AD dataset show that SPD consistently improves these contrastive pre-training baselines and even the supervised pre-training. For example, SPD improves Area Under the Precision-Recall curve (AU-PR) for anomaly segmentation by 5.9% and 6.8% over SimSiam and supervised pre-training respectively in the 2-class high-shot regime. We open-source the project at http://github.com/amazon-research/spot-diff .
Forward citations
Cited by 7 Pith papers
-
SiM3D: Single-instance Multiview Multimodal and Multisetup 3D Anomaly Detection Benchmark
SiM3D provides a multiview, multimodal 3D anomaly detection benchmark with single-instance training and synthetic-to-real evaluation, showing adapted 2D methods often beat multimodal 3D methods on the new voxel-based task.
-
LDU-Bench: Multimodal LLM Evaluation for Lithography Defect Understanding under Layout-Varying Circuit Backgrounds
LDU-Bench evaluates multimodal LLMs on four lithography review tasks and finds a systematic drop-off from defect triage to morphology, localization, and cause analysis.
-
Self-Navigated Residual Mamba for Universal Industrial Anomaly Detection
SNARM combines memory-bank residuals, self-referential in-image residuals, and residual-guided Mamba scanning to report state-of-the-art anomaly detection scores on three benchmarks.
-
LAD-Reasoner: Tiny Multimodal Models are Good Reasoners for Logical Anomaly Detection
A 3B multimodal model fine-tuned with supervised learning and GRPO achieves accuracy and F1 comparable to a 72B model on logical anomaly detection while generating structured textual explanations.
-
Multimodal Task Representation Memory Bank vs. Catastrophic Forgetting in Anomaly Detection
By storing per-task key-prompt-multimodal knowledge and applying structure-based contrastive learning, MTRMB reports higher continual anomaly detection accuracy and lower forgetting than prior methods on MVTec AD and VisA.
-
MoViAD: A Modular Library for Visual Anomaly Detection
A modular visual anomaly detection library is described, but without code, benchmarks, or experimental validation of its capabilities.
-
Friend or Foe? Harnessing Controllable Overfitting for Anomaly Detection
COAD adds a controlled overfitting stage to student-teacher anomaly detectors and reports modest AUROC gains, but its new metric RADI is just AUROC and the 'golden overfitting interval' is fitted, not derived.
Discussion (0). Continue with ORCID to comment.