Abstract: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 <a class="link-external link-https" href="https://github.com/HanboBizl/DMNet" rel="external noopener nofollow">this https URL</a>.

Learn to Few-Shot Segment Remote Sensing Images from Irrelevant Data

Not Just Learning from Others but Relying on Yourself: A New Perspective on Few-Shot Segmentation in Remote Sensing

Iterative Few-shot Semantic Segmentation from Image Label Text

Few-Shot Aerial Image Semantic Segmentation Leveraging Pyramid Correlation Fusion

Pixel Matching Network for Cross-Domain Few-Shot Segmentation

Global Rectification and Decoupled Registration for Few-Shot Segmentation in Remote Sensing Imagery.

Multi-Similarity Enhancement Network for Few-Shot Segmentation.

Few-Shot Rotation-Invariant Aerial Image Semantic Segmentation

Holistic Mutual Representation Enhancement for Few-Shot Remote Sensing Segmentation

Adaptive FSS: A Novel Few-Shot Segmentation Framework Via Prototype Enhancement

Pixel-Wise and Class-Wise Semantic Cues for Few-Shot Segmentation in Astronaut Working Scenes

Deep Reasoning Network for Few-shot Semantic Segmentation

Cross-Domain Few-Shot Semantic Segmentation via Doubly Matching Transformation

Mining Latent Classes for Few-shot Segmentation

Reflection Invariance Learning for Few-shot Semantic Segmentation

Disentangled Foreground-Semantic Adapter Network for Generalized Aerial Image Few-Shot Semantic Segmentation

Deep Updated Subspace Networks for Few-Shot Remote Sensing Scene Classification

ICSFF: Information Constraint on Self-Supervised Feature Fusion for Few-Shot Remote Sensing Image Classification

Embedding Generalized Semantic Knowledge into Few-Shot Remote Sensing Segmentation

Cross-Domain Few-Shot Segmentation via Iterative Support-Query Correspondence Mining

Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training