Chengkun Wang

OPERA: Omni-Supervised Representation Learning with Hierarchical Supervisions

2022 · arXiv (Cornell University)

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 …

OPERA: Omni-Supervised Representation Learning with Hierarchical Supervisions

2023

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 …