Learning by comparing images improved by understanding objects
Learning by comparing images improved by understanding objects
Unsupervised Object-Level Representation Learning from Scene Images
arXiv paper abstract https://arxiv.org/abs/2106.11952
arXiv PDF paper https://arxiv.org/pdf/2106.11952.pdf
Project Web site https://www.mmlab-ntu.com/project/orl/
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.
... infeasible when pre-trained on more complex scene images with many objects.
... introduce Object-level Representation Learning (ORL) ... towards scene images.
... 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.
... 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 ... performance when more unlabeled scene images are available,
... great potential of harnessing unlabeled data in the wild. ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments