{"@context":"https://schema.org","@type":"CreativeWork","@id":"https://froggit.ai/public/capsules/8a6cb941-9194-4ffd-a628-4a69c6248bd9","identifier":"8a6cb941-9194-4ffd-a628-4a69c6248bd9","url":"https://froggit.ai/public/capsules/8a6cb941-9194-4ffd-a628-4a69c6248bd9","name":"AD4AD: Benchmarking Visual Anomaly Detection Models for Safer Autonomous Driving","text":"# AD4AD: Benchmarking Visual Anomaly Detection Models for Safer Autonomous Driving\n\nSource-backed public reference for arXiv:2604.15291.\n\n**Authors:** Fabrizio Genilotti, Arianna Stropeni, Gionata Grotto, Francesco Borsatti, Manuel Barusco, Davide Dalle Pezze, Gian Antonio Susto\n**Primary source:** https://arxiv.org/abs/2604.15291\n**Published:** 2026-04-16T17:54:53Z\n**Updated:** 2026-04-16T17:54:53Z\n**Categories:** cs.CV, cs.AI\n\n## Abstract Summary\nThe reliability of a machine vision system for autonomous driving depends heavily on its training data distribution. When a vehicle encounters significantly different conditions, such as atypical obstacles, its perceptual capabilities can degrade substantially. Unlike many domains where errors carry limited consequences, failures in autonomous driving translate directly into physical risk for passengers, pedestrians, and other road users. To address this challenge, we explore Visual Anomaly Detection (VAD) as a solution. VAD enables the identification of anomalous objects not present during training, allowing the system to alert the driver when an unfamiliar situation is detected. Crucially, VAD models produce pixel-level anomaly maps that can guide driver attention to specific regions of concern without requiring any prior assumptions about the nature or form of the hazard. We benchmark eight state-of-the-art VAD methods on AnoVox, the largest synthetic dataset for anomaly detection in autonomous driving. In particular, we evaluate performance across four backbone architectures spanning from large networks to lightweight ones such as MobileNet and DeiT-Tiny. Our results demonstrate that VAD transfers effectively to road scenes. Notably, Tiny-Dinomaly achieves the best...\n\n## Public Use Notes\n- This capsule summarizes the paper's arXiv metadata and abstract; it is not an independent replication or endorsement of the paper's claims.\n- Use it as a cited research reference for discovery, retrieval, and agent context.\n- For c","keywords":["cs.CV","cs.AI"],"about":[],"citation":["https://arxiv.org/abs/2604.15291"],"isPartOf":{"@type":"Dataset","name":"Forge Cascade Knowledge Graph","url":"https://froggit.ai"},"publisher":{"@type":"Organization","name":"Forge Cascade","url":"https://froggit.ai"},"dateCreated":"2026-04-17T06:00:03.786000Z","dateModified":"2026-06-19T03:22:45Z","isBasedOn":"https://arxiv.org/abs/2604.15291","additionalProperty":[{"@type":"PropertyValue","name":"trust_level","value":100},{"@type":"PropertyValue","name":"verification_status","value":"sources_verified"},{"@type":"PropertyValue","name":"provenance_status","value":"valid"},{"@type":"PropertyValue","name":"evidence_level","value":"primary_source"},{"@type":"PropertyValue","name":"content_hash","value":"3a5bbdcda46bb2db99a95b5034ae598a30556f7d697df76fdddf4165ce511a61"}]}