TY - GEN
T1 - CMDS-AD
T2 - 19th European Conference on Computer Vision, ECCV 2026
AU - Cai, Junhao
AU - Chen, Junyu
AU - Zeng, Deyu
AU - Pang, Junhao
AU - Liang, Qiwei
AU - Zhong, Xiaopin
AU - Wu, Zongze
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026
Y1 - 2026
N2 - Few-shot anomaly detection remains challenging due to limited training data. Multi-modal anomaly detection (MAD) offers a viable solution, leveraging 3D geometric cues to enrich 2D RGB representations and compensate for this scarcity. However, existing MAD methods apply spatially uniform feature processing, conflating stable macroscopic structures with high-frequency localized defect signals, exacerbating cross-modal misalignment and inflating false-positive rates. To overcome this, we present CMDS-AD, a Cross-Modal Dual-Stream Anomaly Detection framework. A LoRA-guided diffusion model generates diverse RGB samples to mitigate extreme data scarcity. For 3D normal augmentation, we employ a pre-trained diffusion model as a normal estimator. Crucially, this estimator inherently acts as a non-linear low-pass filter, directly extracting low-frequency normal representations from RGB inputs. This establishes an auxiliary estimated stream of purely low-frequency information, anchoring robust structural templates and assisting the uncompressed real stream, containing coupled high- and low-frequency components, to precisely isolate micro-defects. A Coordinate-Aware Hierarchical Feature Mapper adaptively aligns cross-modal semantics, while a multiplicative scoring mechanism filters modality-specific noise. Under the extreme 1-shot setting, CMDS-AD achieves absolute performance gains of 5.7% (I-AUROC) and 2.0% (AUPRO) on MVTec 3D-AD, alongside 7.7% and 5.6% improvements on EyeCandies, establishing a new state-of-the-art. Code is available at Junhaocai27/CMDS-AD.
AB - Few-shot anomaly detection remains challenging due to limited training data. Multi-modal anomaly detection (MAD) offers a viable solution, leveraging 3D geometric cues to enrich 2D RGB representations and compensate for this scarcity. However, existing MAD methods apply spatially uniform feature processing, conflating stable macroscopic structures with high-frequency localized defect signals, exacerbating cross-modal misalignment and inflating false-positive rates. To overcome this, we present CMDS-AD, a Cross-Modal Dual-Stream Anomaly Detection framework. A LoRA-guided diffusion model generates diverse RGB samples to mitigate extreme data scarcity. For 3D normal augmentation, we employ a pre-trained diffusion model as a normal estimator. Crucially, this estimator inherently acts as a non-linear low-pass filter, directly extracting low-frequency normal representations from RGB inputs. This establishes an auxiliary estimated stream of purely low-frequency information, anchoring robust structural templates and assisting the uncompressed real stream, containing coupled high- and low-frequency components, to precisely isolate micro-defects. A Coordinate-Aware Hierarchical Feature Mapper adaptively aligns cross-modal semantics, while a multiplicative scoring mechanism filters modality-specific noise. Under the extreme 1-shot setting, CMDS-AD achieves absolute performance gains of 5.7% (I-AUROC) and 2.0% (AUPRO) on MVTec 3D-AD, alongside 7.7% and 5.6% improvements on EyeCandies, establishing a new state-of-the-art. Code is available at Junhaocai27/CMDS-AD.
KW - Diffusion Models
KW - Dual-Stream Optimization
KW - Few-Shot Learning
KW - Multi-Modal Anomaly Detection
UR - https://www.scopus.com/pages/publications/105054501586
U2 - 10.1007/978-3-032-37016-7_23
DO - 10.1007/978-3-032-37016-7_23
M3 - 会议稿件
AN - SCOPUS:105054501586
SN - 9783032370150
T3 - Lecture Notes in Computer Science
SP - 431
EP - 448
BT - Computer Vision – ECCV 2026 - 19th European Conference, 2026, Proceedings
A2 - Favaro, Paolo
A2 - Kukelova, Zuzana
A2 - Maki, Atsuto
A2 - Rohrbach, Anna
A2 - Schindler, Konrad
A2 - Tombari, Federico
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 8 September 2026 through 12 September 2026
ER -