Related papers: LoDisc: Learning Global-Local Discriminative Features for Self-Supervised Fine-Grained Visual Recognition

LoDisc: Learning Global-Local Discriminative Features for Self-Supervised Fine-Grained Visual Recognition

URL: http://arxiv.org/abs/2403.04066v1
Date: Wed, 6 Mar 2024 21:36:38 GMT
Title: LoDisc: Learning Global-Local Discriminative Features for Self-Supervised Fine-Grained Visual Recognition
Authors: Jialu Shi, Zhiqiang Wei, Jie Nie, Lei Huang
Abstract summary: We present to incorporate the subtle local fine-grained feature learning into global self-supervised contrastive learning. A novel pretext task called Local Discrimination (LoDisc) is proposed to explicitly supervise self-supervised model's focus towards local pivotal regions. We show that Local Discrimination pretext task can effectively enhance fine-grained clues in important local regions, and the global-local framework further refines the fine-grained feature representations of images.
Score: 18.442966979622717
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Self-supervised contrastive learning strategy has attracted remarkable attention due to its exceptional ability in representation learning. However, current contrastive learning tends to learn global coarse-grained representations of the image that benefit generic object recognition, whereas such coarse-grained features are insufficient for fine-grained visual recognition. In this paper, we 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. Specifically, a novel pretext task called Local Discrimination (LoDisc) is proposed 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. We show that Local Discrimination pretext task can effectively enhance fine-grained clues in important local regions, and the global-local framework further refines the fine-grained feature representations of images. Extensive experimental results on different fine-grained object recognition tasks demonstrate that the proposed method can lead to a decent improvement in different evaluation settings. Meanwhile, the proposed method is also effective in general object recognition tasks.