Analyzing the effectiveness of image augmentations for face recognition from limited data
This work addresses the challenge of data scarcity in face recognition, but it is incremental as it focuses on analyzing existing augmentation methods rather than introducing new ones.
The paper tackled the problem of improving face recognition systems with limited data by analyzing various image augmentation techniques, finding that augmentations generally enhance quality and that combining generative and basic methods yields the best performance.
This work presents an analysis of the efficiency of image augmentations for the face recognition problem from limited data. We considered basic manipulations, generative methods, and their combinations for augmentations. Our results show that augmentations, in general, can considerably improve the quality of face recognition systems and the combination of generative and basic approaches performs better than the other tested techniques.