UFM: A Simple Path towards Unified Dense Correspondence with Flow

TL;DR

UFM introduces a unified transformer-based model trained on combined data to improve dense correspondence accuracy and speed across various applications, outperforming specialized methods.

Contribution

The paper presents UFM, a simple, unified transformer model trained on combined data for dense correspondence, outperforming specialized approaches in accuracy and speed.

Findings

UFM is 28% more accurate than state-of-the-art flow methods.

UFM has 62% less error and is 6.7x faster than dense wide-baseline matchers.

Unified training enables superior performance across multiple correspondence tasks.

Abstract

Dense image correspondence is central to many applications, such as visual odometry, 3D reconstruction, object association, and re-identification. Historically, dense correspondence has been tackled separately for wide-baseline scenarios and optical flow estimation, despite the common goal of matching content between two images. In this paper, we develop a Unified Flow & Matching model (UFM), which is trained on unified data for pixels that are co-visible in both source and target images. UFM uses a simple, generic transformer architecture that directly regresses the (u,v) flow. It is easier to train and more accurate for large flows compared to the typical coarse-to-fine cost volumes in prior work. UFM is 28% more accurate than state-of-the-art flow methods (Unimatch), while also having 62% less error and 6.7x faster than dense wide-baseline matchers (RoMa). UFM is the first to demonstrate that unified training can outperform specialized approaches across both domains. This result enables fast, general-purpose correspondence and opens new directions for multi-modal, long-range, and real-time correspondence tasks.