Unlabeled Image
Unlabeled image data presents a significant challenge and opportunity in computer vision. Research focuses on leveraging this abundant resource through semi-supervised and unsupervised learning techniques, often employing methods like contrastive learning, generative adversarial networks (GANs), and self-supervised learning with various architectures including vision transformers and convolutional neural networks. These approaches aim to improve model performance in tasks such as image segmentation, object detection, and classification, reducing the reliance on expensive and time-consuming manual labeling. The efficient utilization of unlabeled images has broad implications for advancing various fields, including medical imaging, robotics, and remote sensing.
Papers
L-DAWA: Layer-wise Divergence Aware Weight Aggregation in Federated Self-Supervised Visual Representation Learning
Yasar Abbas Ur Rehman, Yan Gao, Pedro Porto Buarque de Gusmão, Mina Alibeigi, Jiajun Shen, Nicholas D. Lane
FreeCOS: Self-Supervised Learning from Fractals and Unlabeled Images for Curvilinear Object Segmentation
Tianyi Shi, Xiaohuan Ding, Liang Zhang, Xin Yang
Source-Free Domain Adaptive Fundus Image Segmentation with Class-Balanced Mean Teacher
Longxiang Tang, Kai Li, Chunming He, Yulun Zhang, Xiu Li