Thanks to the development of deep learning, promising image classification and retrieval performance within a set of predefined categories can be achieved without extra object-based bounding boxes or dense keypoint-based annotation information in recent years. However, fine-grained image analysis (FGIA) including image recognition and retrieval within sub-categories are much more challenging problem than general image recognition and retrieval due to the small inter-class and large intra-class variation. Related researches of FGIA technology have been widely and successfully applied in smart retail, intelligent transportation, biodiversity conservation and crime prevention. The task of FGIA targets recognizing the variances among images categorized in subordinate classes, e.g., species of birds, types of cars or species of flowers. The main research subjects of FGIA contain two fundamental fine-grained research areas i.e., fine-grained image recognition and fine-grained image retrieval.

A list of the paper types that you will consider for this SI e.g. research/review/case study/brief. This list should not exceed the paper types normally considered in this journal.