[2411.19888] FlowCLAS: Enhancing Normalizing Flow Via Contrastive Learning For Anomaly Segmentation

Computer Science > Computer Vision and Pattern Recognition arXiv:2411.19888 (cs) [Submitted on 29 Nov 2024 (v1), last revised 4 Mar 2026 (this version, v2)] Title:FlowCLAS: Enhancing Normalizing Flow Via Contrastive Learning For Anomaly Segmentation Authors:Chang Won Lee, Selina Leveugle, Svetlana Stolpner, Chris Langley, Paul Grouchy, Jonathan Kelly, Steven L. Waslander View a PDF of the paper titled FlowCLAS: Enhancing Normalizing Flow Via Contrastive Learning For Anomaly Segmentation, by Chang Won Lee and 6 other authors View PDF HTML (experimental) Abstract:Anomaly segmentation is an essential capability for safety-critical robotics applications that must be aware of unexpected events. Normalizing flows (NFs), a class of generative models, are a promising approach for this task due to their ability to model the inlier data distribution efficiently. However, their performance falters in dynamic scenes, where complex, multi-modal data distributions cause them to struggle with identifying out-of-distribution samples, leaving a performance gap to leading discriminative methods. To address this limitation, we introduce FlowCLAS, a hybrid framework that enhances the traditional maximum likelihood objective of NFs with a discriminative, contrastive loss. Leveraging Outlier Exposure, this objective explicitly enforces a separation between normal and anomalous features in the latent space, retaining the probabilistic foundation of NFs while embedding the discriminative power th...