Learning Dense Feature Matching via Lifting Single 2D Image to 3D Space

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liang, Yingping, Hu, Yutao, Shao, Wenqi, Fu, Ying
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911041064009728
author Liang, Yingping
Hu, Yutao
Shao, Wenqi
Fu, Ying
author_facet Liang, Yingping
Hu, Yutao
Shao, Wenqi
Fu, Ying
contents Feature matching plays a fundamental role in many computer vision tasks, yet existing methods heavily rely on scarce and clean multi-view image collections, which constrains their generalization to diverse and challenging scenarios. Moreover, conventional feature encoders are typically trained on single-view 2D images, limiting their capacity to capture 3D-aware correspondences. In this paper, we propose a novel two-stage framework that lifts 2D images to 3D space, named as \textbf{Lift to Match (L2M)}, taking full advantage of large-scale and diverse single-view images. To be specific, in the first stage, we learn a 3D-aware feature encoder using a combination of multi-view image synthesis and 3D feature Gaussian representation, which injects 3D geometry knowledge into the encoder. In the second stage, a novel-view rendering strategy, combined with large-scale synthetic data generation from single-view images, is employed to learn a feature decoder for robust feature matching, thus achieving generalization across diverse domains. Extensive experiments demonstrate that our method achieves superior generalization across zero-shot evaluation benchmarks, highlighting the effectiveness of the proposed framework for robust feature matching.
format Preprint
id arxiv_https___arxiv_org_abs_2507_00392
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Dense Feature Matching via Lifting Single 2D Image to 3D Space
Liang, Yingping
Hu, Yutao
Shao, Wenqi
Fu, Ying
Computer Vision and Pattern Recognition
Feature matching plays a fundamental role in many computer vision tasks, yet existing methods heavily rely on scarce and clean multi-view image collections, which constrains their generalization to diverse and challenging scenarios. Moreover, conventional feature encoders are typically trained on single-view 2D images, limiting their capacity to capture 3D-aware correspondences. In this paper, we propose a novel two-stage framework that lifts 2D images to 3D space, named as \textbf{Lift to Match (L2M)}, taking full advantage of large-scale and diverse single-view images. To be specific, in the first stage, we learn a 3D-aware feature encoder using a combination of multi-view image synthesis and 3D feature Gaussian representation, which injects 3D geometry knowledge into the encoder. In the second stage, a novel-view rendering strategy, combined with large-scale synthetic data generation from single-view images, is employed to learn a feature decoder for robust feature matching, thus achieving generalization across diverse domains. Extensive experiments demonstrate that our method achieves superior generalization across zero-shot evaluation benchmarks, highlighting the effectiveness of the proposed framework for robust feature matching.
title Learning Dense Feature Matching via Lifting Single 2D Image to 3D Space
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.00392