Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces Zero-OVCD, a two-stage framework for open-vocabulary change detection (OVCD) that operates without requiring pixel-level annotations from the target domain. The first stage generates high-quality pseudo-labels through a combination of candidate-mask refinement, multiscale semantic similarity fusion, and response-guided corrections, significantly improving mask-level semantic discrimination and reducing noise. The second stage trains a change detector using these refined pseudo-labels, achieving substantial F1 score improvements across multiple datasets, thus demonstrating the efficacy of integrating training-free foundation models with noise-aware pseudo-label learning.
Achieving F1 scores above 88% in open-vocabulary change detection without any target-domain pixel-level annotations reveals a groundbreaking approach to tackling noisy data in remote sensing.
Open-vocabulary change detection (OVCD) enables the identification of user-specified land-cover changes in bitemporal remote sensing images, but existing training-free pipelines remain vulnerable to inaccurate candidate masks, ambiguous semantic assignments, and accumulated inference errors. To address these issues, we propose Zero-OVCD, a two-stage framework that requires no pixel-level annotations from the target domain. In the first stage, high-quality change pseudo-labels are generated through complementary candidate-mask refinement, multiscale semantic similarity fusion with margin-based reliability filtering, and response-guided mask correction and completion. These components jointly suppress noisy candidates, enhance mask-level semantic discrimination, and recover missed change regions. In the second stage, a change detector is trained using the generated pseudo-labels, while checkpoint voting and high-agreement sample selection are introduced to mitigate residual pseudo-label noise. On LEVIR-CD, WHU-CD, and S2Looking, Stage I achieves F1 scores of 86.25%, 85.82%, and 50.48%, while Stage II further improves them to 88.65%, 88.85%, and 57.96%, respectively. On SECOND, the macro-average F1 across six category-wise one-vs-rest tasks increases from 47.91% to 50.92%. These results demonstrate that bridging training-free foundation-model inference with noise-aware pseudo-label learning provides an effective solution for open-vocabulary change detection without target-domain pixel-level annotations. Code will be available at https://github.com/1321663019/Zero-OVCD.