mlsanityData to 7 Oct 2026

Back to Point: Exploring Point-Language Models for Zero-Shot 3D Anomaly Detection

arXiv 2603.21511 ↗First posted 23 Mar 2026cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

Zero-shot (ZS) 3D anomaly detection is crucial for reliable industrial inspection, as it enables detecting and localizing defects without requiring any target-category training data. Existing approaches render 3D point clouds into 2D images and leverage pre-trained Vision-Language Models (VLMs) for anomaly detection. However, such strategies inevitably discard geometric details and exhibit limited sensitivity to local anomalies. In this paper, we revisit intrinsic 3D representations and explore the potential of pre-trained Point-Language Models (PLMs) for ZS 3D anomaly detection. We propose BTP (Back To Point), a novel framework that effectively aligns 3D point cloud and textual embeddings. Specifically, BTP aligns multi-granularity patch features with textual representations for localized anomaly detection, while incorporating geometric descriptors to enhance sensitivity to structural anomalies. Furthermore, we introduce a joint representation learning strategy that leverages auxiliary point cloud data to improve robustness and enrich anomaly semantics. Extensive experiments on Real3D-AD and Anomaly-ShapeNet demonstrate that BTP achieves superior performance in ZS 3D anomaly detection. Code will be available at hrefhttps://github.com/wistful-8029/BTP-3DADhttps://github.com/wistful-8029/BTP-3DAD.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 37 of its 59 references that are in the map. First posted on arXiv 23 Mar 2026.

Builds on

37 of 59 references.

References · 37

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  5. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  6. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  7. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  8. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
  9. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 647
  10. 2019Deep Learning for 3D Point Clouds: A Surveycited by 557
  11. 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 311
  12. 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 270
25 more
  1. 2022A Unified Model for Multi-class Anomaly Detectioncited by 267
  2. 2022ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understandingcited by 262
  3. 2022Delving into Out-of-Distribution Detection with Vision-Language Representationscited by 215
  4. 2023ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understandingcited by 199
  5. 2023OpenShape: Scaling Up 3D Shape Representation Towards Open-World Understandingcited by 183
  6. 2023Multimodal Industrial Anomaly Detection via Hybrid Fusioncited by 138
  7. 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 126
  8. 2022Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detectioncited by 92
  9. 2023Real3D-AD: A Dataset of Point Cloud Anomaly Detectioncited by 87
  10. 2023Complementary Pseudo Multimodal Feature for Point Cloud Anomaly Detectioncited by 74
  11. 2025AA-CLIP: Enhancing Zero-shot Anomaly Detection via Anomaly-Aware CLIPcited by 72
  12. 2023Towards Scalable 3D Anomaly Detection and Localization: A Benchmark via 3D Anomaly Synthesis and A Self-Supervised Learning Networkcited by 67
  13. 2023Segment Any Anomaly without Training via Hybrid Prompt Regularizationcited by 59
  14. 2025Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Modelscited by 54
  15. 2024R3D-AD: Reconstruction via Diffusion for 3D Anomaly Detectioncited by 45
  16. 2023EasyNet: An Easy Network for 3D Industrial Anomaly Detectioncited by 43
  17. 2025Bayesian Prompt Flow Learning for Zero-Shot Anomaly Detectioncited by 31
  18. 2024PO3AD: Predicting Point Offsets toward Better 3D Point Cloud Anomaly Detectioncited by 30
  19. 2024PointAD: Comprehending 3D Anomalies from Points and Pixels for Zero-shot 3D Anomaly Detectioncited by 29
  20. 2024Multi-Sensor Object Anomaly Detection: Unifying Appearance, Geometry, and Internal Propertiescited by 26
  21. 2025Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detectioncited by 12
  22. 2024Towards Zero-shot 3D Anomaly Localizationcited by 8
  23. 2026Distilled Large Language Model-Driven Dynamic Sparse Expert Activation Mechanismcited by 7
  24. 2025Exploiting Point-Language Models with Dual-Prompts for 3D Anomaly Detectioncited by 4
  25. 2025MCL-AD: Multimodal Collaboration Learning for Zero-Shot 3D Anomaly Detectioncited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026When Point Clouds Outperform Pixels: Rethinking Zero-Shot Multimodal Anomaly Detectioncited by 0
  2. 2026AT3D-AD: Anomaly Type-Aware 3D Anomaly Detection via Hierarchical Point-Language Alignmentcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Towards Zero-shot Point Cloud Anomaly Detection: A Multi-View Projection Framework
  2. 2026Synthesis4AD: Synthetic Anomalies are All You Need for 3D Anomaly Detection
  3. 2026Towards Generalizable 3D Anomaly Detection via Relational Inconsistency Modeling
  4. 2026Two Steps Are All You Need: Efficient 3D Point Cloud Anomaly Detection with Consistency Models
  5. 2026ZSG-IAD: A Multimodal Framework for Zero-Shot Grounded Industrial Anomaly Detection
  6. 2026TIPS Over Tricks: Simple Prompts for Effective Zero-shot Anomaly Detection
  7. 2024FADE: Few-shot/zero-shot Anomaly Detection Engine using Large Vision-Language Model
  8. 2024VMAD: Visual-enhanced Multimodal Large Language Model for Zero-Shot Anomaly Detection