Deep metric learning for visual understanding: An overview of recent advances

Metric learning aims to learn a distance function to measure the similarity of samples, which plays an important role in many visual understanding applications. Generally, the optimal similarity functions for different visual understanding tasks are task specific because the distributions for data used in different tasks are usually different. It is generally believed that learning a metric from training data can obtain more encouraging performances than handcrafted metrics [1]-[3], e.g., the Euclidean and cosine distances. A variety of metric learning methods have been proposed in the literature [2]-[5], and many of them have been successfully employed in visual understanding tasks such as face recognition [6], [7], image classification [2], [3], visual search [8], [9], visual tracking [10], [11], person reidentification [12], cross-modal matching [13], image set classification [14], and image-based geolocalization [15]-[17].