GS-CLIP：通过几何感知提示词与协同视图表征学习实现零样本 3D 异常检测
GS-CLIP: Zero-shot 3D Anomaly Detection by Geometry-Aware Prompt and Synergistic View Representation Learning
把 3D 几何先验蒸馏进 CLIP 文本提示词，并用渲染图与深度图双流互补融合，为点云零样本异常检测提供了可复用的「几何感知提示词 + 多视图协同」范式，可直接迁移到其他 3D 开放词表任务。
原文依据 dynamically generate text prompts embedded with 3D geometric priors
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论零样本 3D 异常检测（ZS3DAD）是一项新兴任务，旨在不依赖任何目标训练数据的情况下检测目标数据集中的异常，在样本稀缺与数据隐私受限的场景中尤为重要。现有方法通过将 3D 点云投影为 2D 表示来适配 CLIP，但面临若干挑战：投影过程本身会丢失部分几何细节，且对单一 2D 模态的依赖只能提供不完整的视觉理解，限制了其检测多种异常类型的能力。为解决这些局限，我们提出几何感知提示词与协同视图表征学习（GS-CLIP）框架，通过两阶段学习过程使模型能够识别几何异常。在第一阶段，我们动态生成嵌入了 3D 几何先验的文本提示词，这些提示词包含由我们的几何缺陷蒸馏模块（GDDM）蒸馏得到的全局形状上下文与局部缺陷信息。在第二阶段，我们引入协同视图表征学习架构，并行处理渲染图像与深度图像；随后由协同细化模块（SRM）融合两路特征，充分利用二者的互补优势。在四个大规模公开数据集上的综合实验结果表明，GS-CLIP 在检测与分割任务上均取得优越性能。代码将在论文录用后发布。
英文原文摘要 1463 字符
Zero-shot 3D Anomaly Detection (ZS3DAD) is an emerging task that aims to detect anomalies in a target dataset without any target training data, which is particularly important in scenarios constrained by sample scarcity and data privacy concerns. While current methods adapt CLIP by projecting 3D point clouds into 2D representations, they face challenges. The projection inherently loses some geometric details, and the reliance on a single 2D modality provides an incomplete visual understanding, limiting their ability to detect diverse anomaly types. To address these limitations, we propose the Geometry-Aware Prompt and Synergistic View Representation Learning (GS-CLIP) framework, which enables the model to identify geometric anomalies through a two-stage learning process. In the stage 1, we dynamically generate text prompts embedded with 3D geometric priors. These prompts contain global shape context and local defect information distilled by our Geometric Defect Distillation Module (GDDM). In the stage 2, we introduce Synergistic View Representation Learning architecture that processes rendered and depth images in parallel. A Synergistic Refinement Module (SRM) subsequently fuses the features of both streams, capitalizing on their complementary strengths. Comprehensive experimental results on four large-scale public datasets show that GS-CLIP achieves superior performance in detection and segmentation. Code will be released upon acceptance.