mlsanityData to 7 Oct 2026

RoBiS: Robust Binary Segmentation for High-Resolution Industrial Images

arXiv 2505.21152 ↗First posted 27 May 2025cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Robust unsupervised anomaly detection (AD) in real-world scenarios is an important task. Current methods exhibit severe performance degradation on the MVTec AD 2 benchmark due to its complex real-world challenges. To solve this problem, we propose a robust framework RoBiS, which consists of three core modules: (1) Swin-Cropping, a high-resolution image pre-processing strategy to preserve the information of small anomalies through overlapping window cropping. (2) The data augmentation of noise addition and lighting simulation is carried out on the training data to improve the robustness of AD model. We use INP-Former as our baseline, which could generate better results on the various sub-images. (3) The traditional statistical-based binarization strategy (mean+3std) is combined with our previous work, MEBin (published in CVPR2025), for joint adaptive binarization. Then, SAM is further employed to refine the segmentation results. Compared with some methods reported by the MVTec AD 2, our RoBiS achieves a 29.2% SegF1 improvement (from 21.8% to 51.00%) on Test_private and 29.82% SegF1 gains (from 16.7% to 46.52%) on Test_private_mixed. Code is available at https://github.com/xrli-U/RoBiS.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 21 of its 30 references that are in the map. First posted on arXiv 27 May 2025.

Builds on

21 of 30 references.

References · 21

  1. 2023Segment Anythingcited by 8,158
  2. 2023Vision Transformers Need Registerscited by 844
  3. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
  4. 2022Anomaly Detection via Reverse Distillation from One-Class Embeddingcited by 419
  5. 2023SimpleNet: A Simple Network for Image Anomaly Detection and Localizationcited by 256
  6. 2023DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Modelscited by 162
  7. 2023EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latenciescited by 157
  8. 2024RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detectioncited by 116
  9. 2024MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detectioncited by 113
  10. 2024Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detectioncited by 92
  11. 2022DSR -- A dual subspace re-projection network for surface anomaly detectioncited by 89
  12. 2023Stable and low-precision training for large-scale vision-language modelscited by 74
9 more
  1. 2024A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localizationcited by 74
  2. 2024A Survey on Visual Anomaly Detection: Challenge, Approach, and Prospectcited by 69
  3. 2023IM-IAD: Industrial Image Anomaly Detection Benchmark in Manufacturingcited by 64
  4. 2025Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detectioncited by 47
  5. 2023TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detectioncited by 43
  6. 2025The MVTec AD 2 Dataset: Advanced Scenarios for Unsupervised Anomaly Detectioncited by 28
  7. 2023MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detectioncited by 25
  8. 2025Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detectioncited by 11
  9. 2024Dual-level Adaptive Self-Labeling for Novel Class Discovery in Point Cloud Segmentationcited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025From Benchmarks to Reality: Advancing Visual Anomaly Detection by the VAND 3.0 Challengecited by 1
  2. 2026MuRF: Unlocking the Multi-Scale Potential of Vision Foundation Modelscited by 0
  3. 2026SuperADD: Training-free Class-agnostic Anomaly Segmentation -- CVPR 2026 VAND 4.0 Workshop Challenge Industrial Trackcited by 0