Momentum Contrast for Unsupervised Visual Representation Learning

13 Nov 2019Kaiming HeHaoqi FanYuxin WuSaining XieRoss Girshick

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... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Self-Supervised Image Classification ImageNet MoCo (ResNet-50 4x) Top 1 Accuracy 68.6% # 8
Number of Params 375M # 1
Self-Supervised Image Classification ImageNet MoCo (ResNet-50 2x) Top 1 Accuracy 65.4% # 12
Number of Params 94M # 1
Self-Supervised Image Classification ImageNet MoCo (ResNet-50) Top 1 Accuracy 60.6% # 19
Number of Params 24M # 1
Top 1 Accuracy (kNN) 47.1% # 3