Chosen methods of improving small object recognition with weak recognizable features
This addresses the challenge of low sample size and feature representation for small objects in computer vision, but it is incremental as it builds on existing GAN and FasterRCNN techniques.
The paper tackles the problem of small object detection by using GAN-based augmentation to increase data amount and diversity, resulting in improved accuracy on the VOC Pascal dataset compared to traditional augmentation methods.
Many object detection models struggle with several problematic aspects of small object detection including the low number of samples, lack of diversity and low features representation. Taking into account that GANs belong to generative models class, their initial objective is to learn to mimic any data distribution. Using the proper GAN model would enable augmenting low precision data increasing their amount and diversity. This solution could potentially result in improved object detection results. Additionally, incorporating GAN-based architecture inside deep learning model can increase accuracy of small objects recognition. In this work the GAN-based method with augmentation is presented to improve small object detection on VOC Pascal dataset. The method is compared with different popular augmentation strategies like object rotations, shifts etc. The experiments are based on FasterRCNN model.