Fig. 18
From: Attention-guided saptio-temporal feature fusion for robus video surveillance anomaly detection
Comparison of metrics across models: line chart comparing accuracy, precision, recall, and F1-score for EfficientNetV2B0 + HOG, ConvLSTM, and hybrid Model-1 (ResNet-50 + YOLO-v4). Hybrid Model-1 consistently outperforms across most metrics, while EfficientNetV2B0 + HOG excels in recall.
