Unsupervised Single-Scene Semantic Segmentation for Earth Observation

By Sudipan Saha, Muhammad Shahzad, Lichao Mou, Qian Song, Xiao Xiang Zhu. 2022

41 citations recorded by the source corpus.

Earth observation data have huge potential to enrich our knowledge about our planet. An important step in many Earth observation tasks is semantic segmentation. Generally, a large number of pixelwise labeled images are required to train deep models for supervised semantic segmentation. On the contrary, strong intersensor and geographic variations impede the availability of annotated training data in Earth observation. In practice, most Earth observation tasks use only the target scene without assuming availability of any additional scene, labeled or unlabeled. Keeping in mind such constraints, we propose a semantic segmentation method that learns to segment from a single scene, without using any annotation. Earth observation scenes are generally larger than those encountered in typical computer vision datasets. Exploiting this, the proposed method samples smaller unlabeled patches from the scene. For each patch, an alternate view is generated by simple transformations, e.g., addition of noise. Both views are then processed through a two-stream network and weights are iteratively refined using deep clustering, spatial consistency, and contrastive learning in the pixel space. The proposed model automatically segregates the major classes present in the scene and produces the segmentation map. Extensive experiments on four Earth observation datasets collected by different sensors show the effectiveness of the proposed method. Implementation is available at https://gitlab.lrz.de/ai4eo/cd/-/tree/main/unsupContrastiveSemanticSeg.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Continual Barlow Twins: Continual Self-Supervised Learning for Remote Sensing Semantic Segmentation (2022, methodology): “Also in [37], a contrastive method for EO semantic segmentation is proposed.”
FALSE: False Negative Samples Aware Contrastive Learning for Semantic Segmentation of High-Resolution Remote Sensing Image (2022, unknown): “The promise of self-supervised contrastive learning (SSCL) has made it possible to learn the RSI-invariant features from massive unlabeled data [8], [9], [10], [11].”
Self-Supervised Learning in Remote Sensing: A review (2022, methodology): “The authors further integrate DeepCluster, BYOL and MoCo-v2 in the pixel-level to produce segmentation maps [207].”
Self-Supervised Remote Sensing Feature Learning: Learning Paradigms, Challenges, and Future Works (2022, background): “[205], [206], [207] include prototypical contrastive learning (PCL) [200] and Mugs [202].”