[2511.19474] Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks

Computer Science > Computer Vision and Pattern Recognition arXiv:2511.19474 (cs) [Submitted on 22 Nov 2025 (v1), last revised 8 Apr 2026 (this version, v4)] Title:Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks Authors:Jie Li, Hongyi Cai, Mingkang Dong, Muxin Pu, Shan You, Fei Wang, Tao Huang View a PDF of the paper titled Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks, by Jie Li and 6 other authors View PDF HTML (experimental) Abstract:Automatically detecting abnormal events in videos is crucial for modern autonomous systems, yet existing Video Anomaly Detection (VAD) benchmarks lack the scene diversity, balanced anomaly coverage, and temporal complexity needed to reliably assess real-world performance. Meanwhile, the community is increasingly moving toward Video Anomaly Understanding (VAU), which requires deeper semantic and causal reasoning but remains difficult to benchmark due to the heavy manual annotation effort it demands. In this paper, we introduce Pistachio, a new VAD/VAU benchmark constructed entirely through a controlled, generation-based pipeline. By leveraging recent advances in video generation models, Pistachio provides precise control over scenes, anomaly types, and temporal narratives, effectively eliminating the biases and limitations of Internet-collected datasets. Our pipeline integrates scene-conditioned anomaly assignment, multi-step storyline generation, and a temporally consistent long...