Not Just Learning from Others but Relying on Yourself: A New Perspective on Few-Shot Segmentation in Remote Sensing
This work solves the challenge of generalizing few-shot segmentation to remote sensing scenes, which is crucial for applications like land cover analysis, but it is incremental as it builds on existing FSS paradigms with specific improvements.
The paper tackles the problem of few-shot segmentation in remote sensing by addressing extreme intra-class variation and multi-class co-existence, resulting in a new method that achieves state-of-the-art mIoU scores of 49.58% and 51.34% on the iSAID dataset under 1-shot and 5-shot settings, outperforming previous methods by 1.8% and 1.12% respectively.
Few-shot segmentation (FSS) is proposed to segment unknown class targets with just a few annotated samples. Most current FSS methods follow the paradigm of mining the semantics from the support images to guide the query image segmentation. However, such a pattern of `learning from others' struggles to handle the extreme intra-class variation, preventing FSS from being directly generalized to remote sensing scenes. To bridge the gap of intra-class variance, we develop a Dual-Mining network named DMNet for cross-image mining and self-mining, meaning that it no longer focuses solely on support images but pays more attention to the query image itself. Specifically, we propose a Class-public Region Mining (CPRM) module to effectively suppress irrelevant feature pollution by capturing the common semantics between the support-query image pair. The Class-specific Region Mining (CSRM) module is then proposed to continuously mine the class-specific semantics of the query image itself in a `filtering' and `purifying' manner. In addition, to prevent the co-existence of multiple classes in remote sensing scenes from exacerbating the collapse of FSS generalization, we also propose a new Known-class Meta Suppressor (KMS) module to suppress the activation of known-class objects in the sample. Extensive experiments on the iSAID and LoveDA remote sensing datasets have demonstrated that our method sets the state-of-the-art with a minimum number of model parameters. Significantly, our model with the backbone of Resnet-50 achieves the mIoU of 49.58% and 51.34% on iSAID under 1-shot and 5-shot settings, outperforming the state-of-the-art method by 1.8% and 1.12%, respectively. The code is publicly available at https://github.com/HanboBizl/DMNet.