T-Rex Label

Outlier Detection

In the realm of computer vision, outlier detection is the process of singling out data points or objects that deviate remarkably from the majority of data points within a specific dataset or image. These outliers can stem from diverse factors. For example, sensor noise might lead to abnormal data; errors during image acquisition or processing could also create outliers; moreover, the appearance of extraordinary objects in the scene is another potential cause.