Kavli Affiliate: Zheng Zhu | First 5 Authors: Chengkun Wang, Wenzhao Zheng, Zheng Zhu, Jie Zhou, Jiwen Lu | Summary: The pretrain-finetune paradigm in modern computer vision facilitates the success of self-supervised learning, which tends to achieve better transferability than supervised learning. However, with the availability of massive labeled data, a natural question emerges: how […]
Continue.. OPERA: Omni-Supervised Representation Learning with Hierarchical Supervisions