Unsupervised Semantic Correspondence Using Stable Diffusion

Published: 21 Sept 2023, Last Modified: 19 Jan 2024NeurIPS 2023 posterEveryoneRevisionsBibTeX
Keywords: Semantic Correspondence, Stable Diffusion, Optimization-based Inference
TL;DR: We optimize the text embeddings of a pre-trained Stable Diffusion network to estimate semantic correspondences without any supervision.
Abstract: Text-to-image diffusion models are now capable of generating images that are often indistinguishable from real images. To generate such images, these models must understand the semantics of the objects they are asked to generate. In this work we show that, without any training, one can leverage this semantic knowledge within diffusion models to find semantic correspondences – locations in multiple images that have the same semantic meaning. Specifically, given an image, we optimize the prompt embeddings of these models for maximum attention on the regions of interest. These optimized embeddings capture semantic information about the location, which can then be transferred to another image. By doing so we obtain results on par with the strongly supervised state of the art on the PF-Willow dataset and significantly outperform (20.9% relative for the SPair-71k dataset) any existing weakly- or unsupervised method on PF-Willow, CUB-200 and SPair-71k datasets.
Supplementary Material: pdf
Submission Number: 1343