Research map / Label-efficient and robust learning / OOD and anomaly detection
Dynamic Local Aggregation Network with Adaptive Clusterer for Anomaly Detection
Zhiwei Yang, Peng Wu, Jing Liu, Xiaotao Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add DLAN to the autoencoder (ablation) | AE · Ped2 | frame-level AUC: 95.1% -> 97.1% | Improved |
| add DLAN to the autoencoder (ablation) | AE · Avenue | frame-level AUC: 83.9% -> 89.2% | Improved |
| add DLAN to the autoencoder (ablation) | AE · ShanghaiTech | frame-level AUC: 66.7% -> 74.0% | Improved |
| add adaptive clusterer AC instead of fixed 10 prototypes (ablation) | DLAN-AC · Ped2 | frame-level AUC: 97.1% -> 97.6% | Improved |
| add adaptive clusterer AC instead of fixed 10 prototypes (ablation) | DLAN-AC · Avenue | frame-level AUC: 89.2% -> 89.9% | Improved |
| add adaptive clusterer AC instead of fixed 10 prototypes (ablation) | DLAN-AC · ShanghaiTech | frame-level AUC: 74.0% -> 74.7% | Improved |
| add dynamic redundant clustering strategy DRCS (ablation) | DLAN-AC · Ped2 | frame-level AUC: 97.2% -> 97.6% | Improved |
| add dynamic redundant clustering strategy DRCS (ablation) | DLAN-AC · Avenue | frame-level AUC: 89.4% -> 89.9% | Improved |
| add dynamic redundant clustering strategy DRCS (ablation) | DLAN-AC · ShanghaiTech | frame-level AUC: 74.1% -> 74.7% | Improved |
| initialize DLAN centers from AC output instead of random (ablation) | DLAN · Ped2 (10th epoch) | feature aggregation compactness | Improved |
| replace memory-bank prototypes (MPN) with DLAN-AC | DLAN-AC vs MPN · ShanghaiTech | frame-level AUC: 73.8% -> 74.7% | Improved |
| replace ROADMAP with DLAN-AC | DLAN-AC vs ROADMAP · ShanghaiTech | frame-level AUC: 76.6% -> 74.7% | Hurt |
Closest papers by meaning
- A Comprehensive Augmentation Framework for Anomaly Detection (AAAI 2024 · similarity 0.83)
- COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection (NeurIPS 2026 · similarity 0.82)
- Towards a Unified Framework of Clustering-based Anomaly Detection (ICML 2025 · similarity 0.82)
- PANDA: Towards Generalist Video Anomaly Detection via Agentic AI Engineer (NeurIPS 2025 · similarity 0.82)
- Graph Mixture of Experts and Memory-augmented Routers for Multivariate Time Series Anomaly Detection (AAAI 2025 · similarity 0.82)
- UCF-Crime-DVS: A Novel Event-Based Dataset for Video Anomaly Detection with Spiking Neural Networks (AAAI 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
