Online Monitoring Framework for Automotive Time Series Data using JEPA Embeddings
Not on the map. Why?
Abstract (arXiv)
As autonomous vehicles are rolled out, measures must be taken to ensure their safe operation. In order to supervise a system that is already in operation, monitoring frameworks are frequently employed. These run continuously online in the background, supervising the system status and recording anomalies. This work proposes an online monitoring framework to detect anomalies in object state representations. Thereby, a key challenge is creating a framework for anomaly detection without anomaly labels, which are usually unavailable for unknown anomalies. To address this issue, this work applies a self-supervised embedding method to translate object data into a latent representation space. For this, a JEPA-based self-supervised prediction task is constructed, allowing training without anomaly labels and the creation of rich object embeddings. The resulting expressive JEPA embeddings serve as input for established anomaly detection methods, in order to identify anomalies within object state representations. This framework is particularly useful for applications in real-world environments, where new or unknown anomalies may occur during operation for which there are no labels available. Experiments performed on the publicly available, real-world nuScenes dataset illustrate the framework's capabilities.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 38 references that are in the map. First posted on arXiv 10 Feb 2026.
Builds on
24 of 38 references.
References · 24
- 2017Attention Is All You Needcited by 59,932
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
- 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
- 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 738
- 2024Revisiting Feature Prediction for Learning Visual Representations from Videocited by 490
- 2022VOS: Learning What You Don't Know by Virtual Outlier Synthesiscited by 300
- 2022Open Vocabulary Semantic Segmentation with Patch Aligned Contrastive Learningcited by 115
- 2022Perceptual Grouping in Contrastive Vision-Language Modelscited by 82
- 2024DINOv2 Meets Text: A Unified Framework for Image- and Pixel-Level Vision-Language Alignmentcited by 79
- 2023FocalFormer3D : Focusing on Hard Instance for 3D Object Detectioncited by 58
- 2023MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Featurescited by 53
12 more
- 2024Reliable Trajectory Prediction and Uncertainty Quantification with Conditioned Diffusion Modelscited by 13
- 2024T-JEPA: A Joint-Embedding Predictive Architecture for Trajectory Similarity Computationcited by 11
- 2023LS-VOS: Identifying Outliers in 3D Object Detections Using Latent Space Virtual Outlier Synthesiscited by 7
- 2021Novelty Detection and Analysis of Traffic Scenario Infrastructures in the Latent Space of a Vision Transformer-Based Triplet Autoencodercited by 4
- 2023PEM: Perception Error Model for Virtual Testing of Autonomous Vehiclescited by 4
- 2025Hybrid Machine Learning Model with a Constrained Action Space for Trajectory Predictioncited by 3
- 2025Adaptive Neural Networks for Intelligent Data-Driven Developmentcited by 3
- 2025Assessing the Completeness of Traffic Scenario Categories for Automated Highway Driving Functions via Cluster-based Analysiscited by 3
- 2022ExAgt: Expert-guided Augmentation for Representation Learning of Traffic Scenarioscited by 1
- 2023Conditioning Latent-Space Clusters for Real-World Anomaly Classificationcited by 1
- 2024Label-Free Model Failure Detection for Lidar-based Point Cloud Segmentationcited by 1
- 2025Inconsistency-based Active Learning for LiDAR Object Detectioncited by 1
Built on by
Not cited yet.