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. Project page: https://www.mmlab-ntu.com/project/orl/.
翻译:自我监督的自我监督学习在很大程度上缩小了在图像网络上接受监督前培训的差距。 但是,它的成功在很大程度上依赖于图像网络的物体中心前端, 也就是说, 对同一图像的不同增强的视图与同一对象相对应。 当对许多天体的更复杂的场景图像进行预先培训时, 这样的大量调整的限制立即变得不可行。 为了克服这一限制, 我们引入了目标级别代表学习( ORL ), 一个针对现场图像的新的自我监督学习框架。 我们的关键洞察力是利用图像层面的自我监督前端培训前端作为发现目标级别语义通信的先端, 从而实现从场景图像中进行目标层面的代表学习。 COCOCO的大规模实验显示, ORL 大大改进了在现场图像上自我监督学习的性能, 甚至超越了监督的图像网络前方对若干下游任务的培训。 此外, ORL 改进了下游的性能, 展示了在野外使用未加标签的图像的巨大潜力。 我们希望我们的方法能够激励未来对更通用的图像进行研究: http/ Propervidustrationalpulations: http/ produstrublistrualmentalmental