Bridging Supervision Gaps: A Unified Framework for Remote Sensing Change Detection

2026-01-25

AI Summary

This work addresses the challenges of high pixel-level annotation costs and the difficulty of unifying heterogeneous supervision signals in remote sensing change detection. The authors propose UniCD, a unified change detection framework that, for the first time, integrates supervised, weakly supervised, and unsupervised learning within a single collaborative architecture. By leveraging a shared encoder and a multi-branch mechanism, UniCD enables deep fusion of diverse supervision signals. Key innovations include a spatiotemporal awareness module (STAM), change representation regularization (CRR), and semantic prior–driven change inference (SPCI), which collectively bridge the supervision gap. Experiments demonstrate that UniCD achieves state-of-the-art performance across all three settings, outperforming existing methods by 12.72% and 12.37% under weakly supervised and unsupervised protocols, respectively, on benchmark datasets such as LEVIR-CD.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Remote Sensing / Geospatial AI
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Change detection (CD) aims to identify surface changes from multi-temporal remote sensing imagery. In real-world scenarios, Pixel-level change labels are expensive to acquire, and existing models struggle to adapt to scenarios with diverse annotation availability. To tackle this challenge, we propose a unified change detection framework (UniCD), which collaboratively handles supervised, weakly-supervised, and unsupervised tasks through a coupled architecture. UniCD eliminates architectural barriers through a shared encoder and multi-branch collaborative learning mechanism, achieving deep coupling of heterogeneous supervision signals. Specifically, UniCD consists of three supervision-specific branches. In the supervision branch, UniCD introduces the spatial-temporal awareness module (STAM), achieving efficient synergistic fusion of bi-temporal features. In the weakly-supervised branch, we construct change representation regularization (CRR), which steers model convergence from coarse-grained activations toward coherent and separable change modeling. In the unsupervised branch, we propose semantic prior-driven change inference (SPCI), which transforms unsupervised tasks into controlled weakly-supervised path optimization. Experiments on mainstream datasets demonstrate that UniCD achieves optimal performance across three tasks. It exhibits significant accuracy improvements in weakly and unsupervised scenarios, surpassing current state-of-the-art by 12.72% and 12.37% on LEVIR-CD, respectively.

Problem

• Change Detection
• Remote Sensing
• Weakly-supervised Learning
• Unsupervised Learning
• Label Scarcity

Innovation

• Unified Framework
• Change Detection
• Weakly-supervised Learning
• Unsupervised Learning
• Multi-branch Collaborative Learning