MINIMA: Modality Invariant Image Matching

TL;DR

MINIMA introduces a scalable data generation approach and a new dataset to improve universal image matching across multiple modalities, significantly outperforming existing methods in diverse cross-modal scenarios.

Contribution

The paper presents a unified framework and a large-scale multimodal dataset, enabling improved generalization and performance in cross-modal image matching tasks.

Findings

MINIMA outperforms baseline methods in in-domain and zero-shot cross-modal matching.

The generated MD-syn dataset effectively transfers RGB data diversity to multimodal matching.

The approach achieves superior results across 19 different cross-modal cases.

Abstract

Image matching for both cross-view and cross-modality plays a critical role in multimodal perception. In practice, the modality gap caused by different imaging systems/styles poses great challenges to the matching task. Existing works try to extract invariant features for specific modalities and train on limited datasets, showing poor generalization. In this paper, we present MINIMA, a unified image matching framework for multiple cross-modal cases. Without pursuing fancy modules, our MINIMA aims to enhance universal performance from the perspective of data scaling up. For such purpose, we propose a simple yet effective data engine that can freely produce a large dataset containing multiple modalities, rich scenarios, and accurate matching labels. Specifically, we scale up the modalities from cheap but rich RGB-only matching data, by means of generative models. Under this setting, the matching labels and rich diversity of the RGB dataset are well inherited by the generated multimodal data. Benefiting from this, we construct MD-syn, a new comprehensive dataset that fills the data gap for general multimodal image matching. With MD-syn, we can directly train any advanced matching pipeline on randomly selected modality pairs to obtain cross-modal ability. Extensive experiments on in-domain and zero-shot matching tasks, including $19$ cross-modal cases, demonstrate that our MINIMA can significantly outperform the baselines and even surpass modality-specific methods. The dataset and code are available at https://github.com/LSXI7/MINIMA.