mlsanityData to 7 Oct 2026

SAM Struggles in Concealed Scenes -- Empirical Study on Segment Anything

arXiv 2304.06022 ↗First posted 12 Apr 2023cs.CVcited by 75

Not on the map. Why?

Abstract (arXiv)

Segmenting anything is a ground-breaking step toward artificial general intelligence, and the Segment Anything Model (SAM) greatly fosters the foundation models for computer vision. We could not be more excited to probe the performance traits of SAM. In particular, exploring situations in which SAM does not perform well is interesting. In this report, we choose three concealed scenes, i.e., camouflaged animals, industrial defects, and medical lesions, to evaluate SAM under unprompted settings. Our main observation is that SAM looks unskilled in concealed scenes.

Key facts. Cited by 75 papers in the map (data through 7 Oct 2026). Builds on 16 of its 23 references that are in the map. First posted on arXiv 12 Apr 2023.

Builds on

16 of 23 references.

References · 16

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  3. 2023Segment Anythingcited by 8,158
  4. 2021PVT v2: Improved Baselines with Pyramid Vision Transformercited by 749
  5. 2021The Medical Segmentation Decathloncited by 620
  6. 2021The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classificationcited by 507
  7. 2017Structure-measure: A New Way to Evaluate Foreground Mapscited by 480
  8. 2021Anabranch Network for Camouflaged Object Segmentationcited by 227
  9. 2021Concealed Object Detectioncited by 212
  10. 2021Simultaneously Localize, Segment and Rank the Camouflaged Objectscited by 198
  11. 2020Inf-Net: Automatic COVID-19 Lung Infection Segmentation from CT Imagescited by 184
  12. 2019Segmentation-Based Deep-Learning Approach for Surface-Defect Detectioncited by 145
4 more
  1. 2021Salient Object Detection via Integrity Learningcited by 133
  2. 2015Salient Object Detection: A Benchmarkcited by 130
  3. 2022High-resolution Iterative Feedback Network for Camouflaged Object Detectioncited by 78
  4. 2022CamoFormer: Masked Separable Attention for Camouflaged Object Detectioncited by 58

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 34; 2024: 27; 2025: 10; 2026: 4.

75 papers.

Sort:

By area:

  1. 2023Customized Segment Anything Model for Medical Image Segmentationcited by 277
  2. 2023Segment Anything in Medical Imagescited by 226
  3. 2023SAM-Med2Dcited by 123
  4. 2023SAM-6D: Segment Anything Model Meets Zero-Shot 6D Object Pose Estimationcited by 115
  5. 2024Segment Anything Model for Medical Image Segmentation: Current Applications and Future Directionscited by 97
  6. 2023Weakly-Supervised Concealed Object Segmentation with SAM-based Pseudo Labeling and Multi-scale Feature Groupingcited by 92
  7. 2023A Comprehensive Survey on Segment Anything Model for Vision and Beyondcited by 64
  8. 2023Segment Anything in 3D with Radiance Fieldscited by 61
  9. 2023Can SAM Segment Polyps?cited by 59
  10. 2023Input Augmentation with SAM: Boosting Medical Image Segmentation with Segmentation Foundation Modelcited by 56
  11. 2023SA-Med2D-20M Dataset: Segment Anything in 2D Medical Imaging with 20 Million maskscited by 54
  12. 2023Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical Segmentationcited by 51
38 more
  1. 2024Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decodingcited by 48
  2. 2023Advances in Deep Concealed Scene Understandingcited by 47
  3. 2023A Survey on Segment Anything Model (SAM): Vision Foundation Model Meets Prompt Engineeringcited by 45
  4. 2023Annotation-free Audio-Visual Segmentationcited by 44
  5. 2023SAM3D: Zero-Shot 3D Object Detection via Segment Anything Modelcited by 40
  6. 2023Relax Image-Specific Prompt Requirement in SAM: A Single Generic Prompt for Segmenting Camouflaged Objectscited by 40
  7. 2023Segment Anything Model for Medical Images?cited by 39
  8. 2024A Survey of Camouflaged Object Detection and Beyondcited by 37
  9. 2024SAGD: Boundary-Enhanced Segment Anything in 3D Gaussian via Gaussian Decompositioncited by 29
  10. 2024Effectiveness Assessment of Recent Large Vision-Language Modelscited by 29
  11. 2024Foundation Models for Biomedical Image Segmentation: A Surveycited by 28
  12. 2023Segment Anything Model (SAM) for Radiation Oncologycited by 24
  13. 2024Segment Any Cell: A SAM-based Auto-prompting Fine-tuning Framework for Nuclei Segmentationcited by 24
  14. 2023An Empirical Study on the Robustness of the Segment Anything Model (SAM)cited by 23
  15. 2023SAM-PARSER: Fine-tuning SAM Efficiently by Parameter Space Reconstructioncited by 23
  16. 2023Towards Segment Anything Model (SAM) for Medical Image Segmentation: A Surveycited by 21
  17. 2024Exploring Deeper! Segment Anything Model with Depth Perception for Camouflaged Object Detectioncited by 20
  18. 2024Spider: A Unified Framework for Context-dependent Concept Segmentationcited by 18
  19. 2023Any-to-Any Style Transfer: Making Picasso and Da Vinci Collaboratecited by 17
  20. 2024Interactive 3D Medical Image Segmentation with SAM 2cited by 17
  21. 2023Knowledge distillation with Segment Anything (SAM) model for Planetary Geological Mappingcited by 16
  22. 2024OmniCount: Multi-label Object Counting with Semantic-Geometric Priorscited by 16
  23. 2024BA-SAM: Scalable Bias-Mode Attention Mask for Segment Anything Modelcited by 15
  24. 2023Leaf Only SAM: A Segment Anything Pipeline for Zero-Shot Automated Leaf Segmentationcited by 12
  25. 2024ASAM: Boosting Segment Anything Model with Adversarial Tuningcited by 12
  26. 2025Self-Prompt SAM: Medical Image Segmentation via Automatic Prompt SAM Adaptationcited by 12
  27. 2024Explicit Motion Handling and Interactive Prompting for Video Camouflaged Object Detectioncited by 11
  28. 2023Temporally-Extended Prompts Optimization for SAM in Interactive Medical Image Segmentationcited by 9
  29. 2023Large Model Based Referring Camouflaged Object Detectioncited by 8
  30. 2024Inspiring the Next Generation of Segment Anything Models: Comprehensively Evaluate SAM and SAM 2 with Diverse Prompts Towards Context-Dependent Concepts under Different Scenescited by 8
  31. 2023When SAM Meets Shadow Detectioncited by 6
  32. 2023SAMedOCT: Adapting Segment Anything Model (SAM) for Retinal OCTcited by 6
  33. 2023Large Scale Foundation Models for Intelligent Manufacturing Applications: A Surveycited by 6
  34. 2024Bounding Box Stability against Feature Dropout Reflects Detector Generalization across Environmentscited by 6
  35. 2024COMPrompter: reconceptualized segment anything model with multiprompt network for camouflaged object detectioncited by 6
  36. 2023AdapterShadow: Adapting Segment Anything Model for Shadow Detectioncited by 5
  37. 2023Rethinking Polyp Segmentation from an Out-of-Distribution Perspectivecited by 4
  38. 2024MaskSAM: Towards Auto-prompt SAM with Mask Classification for Volumetric Medical Image Segmentationcited by 4

50 most cited of 75 All 75 (JSON)