Momentum Contrast for Unsupervised Visual Representation Learning
This work addresses the problem of reducing reliance on labeled data for computer vision tasks, representing a significant advance rather than an incremental improvement.
The paper tackles unsupervised visual representation learning by introducing Momentum Contrast (MoCo), which builds a dynamic dictionary with a queue and moving-averaged encoder to enable large-scale contrastive learning. It achieves competitive results on ImageNet classification and outperforms supervised pre-training in 7 detection/segmentation tasks, sometimes by large margins, suggesting the gap between unsupervised and supervised learning is largely closed.
We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic dictionary with a queue and a moving-averaged encoder. This enables building a large and consistent dictionary on-the-fly that facilitates contrastive unsupervised learning. MoCo provides competitive results under the common linear protocol on ImageNet classification. More importantly, the representations learned by MoCo transfer well to downstream tasks. MoCo can outperform its supervised pre-training counterpart in 7 detection/segmentation tasks on PASCAL VOC, COCO, and other datasets, sometimes surpassing it by large margins. This suggests that the gap between unsupervised and supervised representation learning has been largely closed in many vision tasks.