Timezone: »

 
Poster
Unsupervised Object-Level Representation Learning from Scene Images
Jiahao Xie · Xiaohang Zhan · Ziwei Liu · Yew Soon Ong · Chen Change Loy

Tue Dec 07 04:30 PM -- 06:00 PM (PST) @
in Poster Session 2 »

Contrastive self-supervised learning has largely narrowed the gap to supervised pre-training on ImageNet. However, its success highly relies on the object-centric priors of ImageNet, i.e., different augmented views of the same image correspond to the same object. Such a heavily curated constraint becomes immediately infeasible when pre-trained on more complex scene images with many objects. To overcome this limitation, we introduce Object-level Representation Learning (ORL), a new self-supervised learning framework towards scene images. Our key insight is to leverage image-level self-supervised pre-training as the prior to discover object-level semantic correspondence, thus realizing object-level representation learning from scene images. Extensive experiments on COCO show that ORL significantly improves the performance of self-supervised learning on scene images, even surpassing supervised ImageNet pre-training on several downstream tasks. Furthermore, ORL improves the downstream performance when more unlabeled scene images are available, demonstrating its great potential of harnessing unlabeled data in the wild. We hope our approach can motivate future research on more general-purpose unsupervised representation learning from scene data.

Keywords: Self-Supervised Learning · Representation Learning

Author Information

Jiahao Xie (Nanyang Technological University)
Xiaohang Zhan (Huawei Technologies Ltd.)
Ziwei Liu (Nanyang Technological University)
Yew Soon Ong (Nanyang Technological University)
Chen Change Loy (Nanyang Technological University)

More from the Same Authors