Self-supervised fine-grained image classification by focus on local important regions with LoDisc
Self-supervised fine-grained image classification by focus on local important regions with LoDisc
LoDisc: Learning Global-Local Discriminative Features for Self-Supervised Fine-Grained Visual Recognition
arXiv paper abstract https://arxiv.org/abs/2403.04066
arXiv PDF paper https://arxiv.org/pdf/2403.04066.pdf
Self-supervised contrastive learning strategy has attracted remarkable attention due to its exceptional ability in representation learning.
However, current contrastive learning ... learn global coarse-grained representations ... that benefit generic object recognition ... are insufficient for fine-grained visual recognition.
... present to incorporate the subtle local fine-grained feature learning into global self-supervised contrastive learning through a pure self-supervised global-local fine-grained contrastive learning framework.
... pretext task called Local Discrimination (LoDisc) ... to explicitly supervise self-supervised model's focus towards local pivotal regions which are captured by a simple-but-effective location-wise mask sampling strategy.
... show that Local Discrimination pretext task ... enhance fine-grained clues in important local regions, and the global-local framework further refines the fine-grained feature representations of images.
... method can lead to a decent improvement in different evaluation settings ... is also effective in general object recognition tasks.
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