Revisiting Self-Supervised Visual Representation Learning

• Published in: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 1920 - 1929
• Main Authors: Kolesnikov, Alexander, Zhai, Xiaohua, Beyer, Lucas
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2019
• Subjects: Deep Learning ; Low-level Vision; Representation Learning
• ISSN: 2575-7075
• DOI: 10.1109/CVPR.2019.00202

Unsupervised visual representation learning remains a largely unsolved problem in computer vision research. Among a big body of recently proposed approaches for unsupervised learning of visual representations, a class of self-supervised techniques achieves superior performance on many challenging benchmarks. A large number of the pretext tasks for self-supervised learning have been studied, but other important aspects, such as the choice of convolutional neural networks (CNN), has not received equal attention. Therefore, we revisit numerous previously proposed self-supervised models, conduct a thorough large scale study and, as a result, uncover multiple crucial insights. We challenge a number of common practices in self-supervised visual representation learning and observe that standard recipes for CNN design do not always translate to self-supervised representation learning. As part of our study, we drastically boost the performance of previously proposed techniques and outperform previously published state-of-the-art results by a large margin. We will release the code for reproducing our experiments when the anonymity requirements are lifted.