FlowIt: Global Matching for Optical Flow with Confidence-Guided Refinement

TL;DR

FlowIt introduces a global matching optical flow method using hierarchical transformers and optimal transport, with confidence-guided refinement, achieving state-of-the-art results and strong generalization across datasets.

Contribution

The paper presents a novel global matching architecture with confidence-guided refinement for optical flow, improving robustness and generalization over existing methods.

Findings

Achieves state-of-the-art results on Sintel and KITTI benchmarks.

Demonstrates superior cross-dataset zero-shot generalization.

Effectively models long-range correspondences with hierarchical transformers.

Abstract

We present FlowIt, a novel architecture for optical flow estimation designed to robustly handle large pixel displacements. At its core, FlowIt leverages a hierarchical transformer architecture that captures extensive global context, enabling the model to effectively model long-range correspondences. To overcome the limitations of localized matching, we formulate the flow initialization as an optimal transport problem. This formulation yields a highly robust initial flow field, alongside explicitly derived occlusion and confidence maps. These cues are then seamlessly integrated into a guided refinement stage, where the network actively propagates reliable motion estimates from high-confidence regions into ambiguous, low-confidence areas. Extensive experiments across the Sintel, KITTI, Spring, and LayeredFlow datasets validate the efficacy of our approach. FlowIt achieves state-of-the-art results on the competitive Sintel and KITTI benchmarks, while simultaneously establishing new state-of-the-art cross-dataset zero-shot generalization performance on Sintel, Spring, and LayeredFlow.