Unsupervised Visual Representation Learning by Context Prediction
2015·Arxiv