REDI-Match: Rotation-Equivariant Distillation for Efficient and Robust Dense Matching

2026-06-23

AI Summary

This work addresses the limited robustness of vision foundation models to severe in-plane rotations in dense matching tasks by proposing the REDI-Match framework. It introduces a novel rotation-equivariant distillation paradigm that transfers semantic knowledge from a non-equivariant vision foundation model to a lightweight rotation-equivariant encoder. To resolve global coordinate ambiguities, an entropy-driven spatial alignment module is incorporated. By integrating a rotation-equivariant geometric architecture with discrete rotation hypothesis evaluation, REDI-Match achieves strict rotation equivariance while preserving strong semantic capabilities. Experiments demonstrate that the method sets new state-of-the-art results across multiple benchmarks, improving pose accuracy by 13.89% on the SatAst dataset and achieving an inference speed of 41 FPS—1.9× faster than the current best approach.

Abstract

Vision Foundation Models (VFMs) have significantly advanced dense feature matching, yet severe in-plane rotation remains a critical challenge. Existing solutions face a fundamental dilemma: data-driven methods require inefficient parameter scaling to implicitly learn rotations, whereas strictly equivariant networks lack the semantic capacity of modern VFMs. Consequently, current frameworks typically freeze VFMs and shift the entire burden of rotation generalization to the downstream decoder. To break this architectural bottleneck, we propose REDI-Match, an efficient framework driven by a novel Rotation-Equivariant Distillation (REDI) paradigm. Instead of relying on rotation data augmentation to establish rotational correspondences, REDI distills the non-equivariant semantic representations of a VFM into a lightweight, strictly rotation-equivariant encoder, leveraging an equivariant geometric architecture to constrain robust high-dimensional semantics. To fully exploit these features, we equip the decoder with an entropy-driven spatial alignment module. By evaluating discrete rotation hypotheses, this mechanism explicitly locks onto the canonical coordinate system, eliminating global ambiguity before continuous refinement. Extensive experiments demonstrate that REDI-Match establishes a new state-of-the-art (SOTA) across multiple benchmarks. Notably, it achieves a 13.89% absolute pose accuracy improvement on the highly challenging SatAst dataset while operating 1.9x faster than the current SOTA (RoMa v2), enabling real-time inference (~41 FPS) on a single RTX 4090 GPU. Code: https://github.com/YinjiGe/REDI-Match.

Problem

• dense matching
• in-plane rotation
• rotation equivariance
• vision foundation models
• feature matching

Innovation

• Rotation-Equivariant Distillation
• Dense Matching
• Vision Foundation Models
• Spatial Alignment
• Equivariant Architecture