The field of anomaly detection in computer vision is moving towards more efficient and robust methods, with a focus on real-time detection and handling of complex textures and multimodal data. Recent developments have led to significant improvements in detection accuracy and speed, with some methods achieving state-of-the-art performance across multiple benchmarks. Notably, the introduction of new datasets and frameworks has enabled more realistic evaluations of anomaly detection methods, highlighting the need for more robust and generalizable solutions. Noteworthy papers include: Quantized FCA, which proposes a real-time method for zero-shot texture anomaly detection with a 10x speedup and little to no loss in accuracy. MIRAD, which introduces a comprehensive real-world robust anomaly detection dataset for mass individualization, capturing the complexities of social manufacturing and highlighting the need for more robust quality control solutions. Towards a Generalizable Fusion Architecture for Multimodal Object Detection, which presents a preprocessing architecture designed to enhance the fusion of RGB and infrared inputs, achieving improved performance across different multimodal challenges. One Dinomaly2 Detect Them All, which presents a unified framework for full-spectrum unsupervised anomaly detection, achieving superior performance across multiple modalities, task settings, and application domains.