PatchRot: A Self-Supervised Technique for Training Vision Transformers
This addresses the high cost of labeling data for vision transformers, offering a domain-specific improvement in computer vision.
The paper tackles the problem of vision transformers requiring large labeled datasets by proposing PatchRot, a self-supervised technique that rotates images and patches to predict angles, resulting in learned features that outperform supervised learning and baselines on various datasets.
Vision transformers require a huge amount of labeled data to outperform convolutional neural networks. However, labeling a huge dataset is a very expensive process. Self-supervised learning techniques alleviate this problem by learning features similar to supervised learning in an unsupervised way. In this paper, we propose a self-supervised technique PatchRot that is crafted for vision transformers. PatchRot rotates images and image patches and trains the network to predict the rotation angles. The network learns to extract both global and local features from an image. Our extensive experiments on different datasets showcase PatchRot training learns rich features which outperform supervised learning and compared baseline.