Segment Any Change
It addresses the open problem of zero-shot change detection for computer vision applications, offering a training-free solution that generalizes to unseen change types and data distributions, though it builds incrementally on existing foundation models.
The paper tackles zero-shot change detection in images by proposing AnyChange, a model that adapts the Segment Anything Model (SAM) without training, achieving a new state-of-the-art on the SECOND benchmark with up to 4.4% F1 score improvement and competitive accuracy using minimal manual annotations.
Visual foundation models have achieved remarkable results in zero-shot image classification and segmentation, but zero-shot change detection remains an open problem. In this paper, we propose the segment any change models (AnyChange), a new type of change detection model that supports zero-shot prediction and generalization on unseen change types and data distributions. AnyChange is built on the segment anything model (SAM) via our training-free adaptation method, bitemporal latent matching. By revealing and exploiting intra-image and inter-image semantic similarities in SAM's latent space, bitemporal latent matching endows SAM with zero-shot change detection capabilities in a training-free way. We also propose a point query mechanism to enable AnyChange's zero-shot object-centric change detection capability. We perform extensive experiments to confirm the effectiveness of AnyChange for zero-shot change detection. AnyChange sets a new record on the SECOND benchmark for unsupervised change detection, exceeding the previous SOTA by up to 4.4% F$_1$ score, and achieving comparable accuracy with negligible manual annotations (1 pixel per image) for supervised change detection. Code is available at https://github.com/Z-Zheng/pytorch-change-models.