Self-supervised Deep Learning for Flower Image Segmentation

摘要: Segmentation plays an important role in imagebased plant phenotyping applications. Deep learning has led to a dramatic improvement segmentation performance. Most deep learning-based methods are supervised and require abundant application-specific training data. Considering the wide range of applications, such data may not be always available. To mitigate this problem, we introduce method that exploits power without using any prior training. In paper, specifically focus on flower segmentation. Recurrence information inside image is used train image-specific network subsequently for The proposed self-supervised as it internal statistics input labeled best our knowledge, first unsupervised single-image