Parametric Instance Classification for Unsupervised Visual Feature Learning
This work addresses the complexity and resource demands of unsupervised learning for computer vision researchers, though it is incremental as it builds on existing instance discrimination approaches.
The paper tackles unsupervised visual feature learning by introducing parametric instance classification (PIC), a simpler one-branch framework that matches the effectiveness of state-of-the-art methods like SimCLR and MoCo v2, and proposes techniques to improve efficiency and scalability.
This paper presents parametric instance classification (PIC) for unsupervised visual feature learning. Unlike the state-of-the-art approaches which do instance discrimination in a dual-branch non-parametric fashion, PIC directly performs a one-branch parametric instance classification, revealing a simple framework similar to supervised classification and without the need to address the information leakage issue. We show that the simple PIC framework can be as effective as the state-of-the-art approaches, i.e. SimCLR and MoCo v2, by adapting several common component settings used in the state-of-the-art approaches. We also propose two novel techniques to further improve effectiveness and practicality of PIC: 1) a sliding-window data scheduler, instead of the previous epoch-based data scheduler, which addresses the extremely infrequent instance visiting issue in PIC and improves the effectiveness; 2) a negative sampling and weight update correction approach to reduce the training time and GPU memory consumption, which also enables application of PIC to almost unlimited training images. We hope that the PIC framework can serve as a simple baseline to facilitate future study.