Zero-Shot Image Feature Consensus with Deep Functional Maps

arXiv:2403.12038v1 Announce Type: new
Abstract: Correspondences emerge from large-scale vision models trained for generative and discriminative tasks. This has been revealed and benchmarked by computing correspondence maps between pairs of images, using nearest neighbors on the feature grids. Existing work has attempted to improve the quality of these correspondence maps by carefully mixing features from different sources, such as by combining the features of different layers or networks. We point out that a better correspondence strategy is available, which directly …

abstract arxiv computing consensus cs.cv feature functional generative image images maps neighbors quality scale tasks type vision vision models work zero-shot