VETime: Vision Enhanced Zero-Shot Time Series Anomaly Detection
Not on the map. Why?
Abstract (arXiv)
Time-series anomaly detection (TSAD) requires identifying both immediate Point Anomalies and long-range Context Anomalies. However, existing foundation models face a fundamental trade-off: 1D temporal models provide fine-grained pointwise localization but lack a global contextual perspective, while 2D vision-based models capture global patterns but suffer from information bottlenecks due to a lack of temporal alignment and coarse-grained pointwise detection. To resolve this dilemma, we propose VETime, the first TSAD framework that unifies temporal and visual modalities through fine-grained visual-temporal alignment and dynamic fusion. VETime introduces a Reversible Image Conversion and a Patch-Level Temporal Alignment module to establish a shared visual-temporal timeline, preserving discriminative details while maintaining temporal sensitivity. Furthermore, we design an Anomaly Window Contrastive Learning mechanism and a Task-Adaptive Multi-Modal Fusion to adaptively integrate the complementary perceptual strengths of both modalities. Extensive experiments demonstrate that VETime significantly outperforms state-of-the-art models in zero-shot scenarios, achieving superior localization precision with lower computational overhead than current vision-based approaches. Code available at: https://github.com/yyyangcoder/VETime.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 33 references that are in the map. First posted on arXiv 18 Feb 2026.
Builds on
24 of 33 references.
References · 24
- 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
- 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
- 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
- 2024Chronos: Learning the Language of Time Seriescited by 712
- 2023A decoder-only foundation model for time-series forecastingcited by 650
- 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
- 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 453
- 2018Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholdingcited by 407
- 2024Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Expertscited by 280
- 2024TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysiscited by 113
- 2024VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasterscited by 87
- 2022Local Evaluation of Time Series Anomaly Detection Algorithmscited by 79
12 more
- 2025Time-VLM: Exploring Multimodal Vision-Language Models for Augmented Time Series Forecastingcited by 59
- 2024Towards a General Time Series Anomaly Detector with Adaptive Bottlenecks and Dual Adversarial Decoderscited by 37
- 2025Harnessing Vision Models for Time Series Analysis: A Surveycited by 26
- 2024See it, Think it, Sorted: Large Multimodal Models are Few-shot Time Series Anomaly Analyzerscited by 20
- 2024Position: Quo Vadis, Unsupervised Time Series Anomaly Detection?cited by 17
- 2025Can Multimodal LLMs Perform Time Series Anomaly Detection?cited by 11
- 2025TSPulse: Tiny Pre-Trained Models with Disentangled Representations for Rapid Time-Series Analysiscited by 11
- 2025Vision-Enhanced Time Series Forecasting via Latent Diffusion Modelscited by 9
- 2025Harnessing Vision-Language Models for Time Series Anomaly Detectioncited by 9
- 2025Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancycited by 9
- 2025Teaching Time Series to See and Speak: Forecasting with Aligned Visual and Textual Perspectivescited by 7
- 2025MLLM4TS: Leveraging Vision and Multimodal Language Models for General Time-Series Analysiscited by 4
Built on by
Not cited yet.