Cross-View Completion Models are Zero-shot Correspondence Estimators
In this work, we explore new perspectives on cross-view completion learning by drawing an analogy to self-supervised correspondence learning. Through our analysis, we demonstrate that the cross-attention map within cross-view completion models captures correspondence more effectively than other correlations derived from encoder or decoder features. We verify the effectiveness of the cross-attention map by evaluating on both zero-shot matching and learning-based geometric matching and multi-frame depth estimation. Project page is available at https://cvlab-kaist.github.io/ZeroCo/.
Code (0)
등록된 구현이 없습니다.
Tasks
DecoderDepth EstimationGeometric MatchingSimilar Papers 제목 키워드 기반
Improving Zero-Shot Object-Level Change Detection by Incorporating Visual Correspondence
Detecting object-level changes between two images across possibly different views is a core task in many applications that involve visual inspection or camera surveillance. Existing change-detection approaches suffer fro…
Change DetectionZero-shot GeneralizationZeroReg: Zero-Shot Point Cloud Registration with Foundation Models
State-of-the-art 3D point cloud registration methods rely on labeled 3D datasets for training, which limits their practical applications in real-world scenarios and often hinders generalization to unseen scenes. Leveragi…
DecoderGraph MatchingObjectObject Localization+1ZeroPS: High-quality Cross-modal Knowledge Transfer for Zero-Shot 3D Part Segmentation
Zero-shot 3D part segmentation is a challenging and fundamental task. In this work, we propose a novel pipeline, ZeroPS, which achieves high-quality knowledge transfer from 2D pretrained foundation models (FMs), SAM and …
3D Part SegmentationInstance SegmentationSemantic SegmentationTransfer Learning+1Can Visual Foundation Models Achieve Long-term Point Tracking?
Large-scale vision foundation models have demonstrated remarkable success across various tasks, underscoring their robust generalization capabilities. While their proficiency in two-view correspondence has been explored,…
Point TrackingUnlocking Zero-shot Potential of Semi-dense Image Matching via Gaussian Splatting
Learning-based image matching critically depends on large-scale, diverse, and geometrically accurate training data. 3D Gaussian Splatting (3DGS) enables photorealistic novel-view synthesis and thus is attractive for data…
Image Matching