Saved in:
Bibliographic Details
Main Authors: Zhang, Kaichen, Sheng, Tianxiang, Shi, Xuanming
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.03715
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912746296049664
author Zhang, Kaichen
Sheng, Tianxiang
Shi, Xuanming
author_facet Zhang, Kaichen
Sheng, Tianxiang
Shi, Xuanming
contents This paper presents DINO-RotateMatch, a deep-learning framework designed to address the chal lenges of image matching in large-scale 3D reconstruction from unstructured Internet images. The method integrates a dataset-adaptive image pairing strategy with rotation-aware keypoint extraction and matching. DINO is employed to retrieve semantically relevant image pairs in large collections, while rotation-based augmentation captures orientation-dependent local features using ALIKED and Light Glue. Experiments on the Kaggle Image Matching Challenge 2025 demonstrate consistent improve ments in mean Average Accuracy (mAA), achieving a Silver Award (47th of 943 teams). The results confirm that combining self-supervised global descriptors with rotation-enhanced local matching offers a robust and scalable solution for large-scale 3D reconstruction.
format Preprint
id arxiv_https___arxiv_org_abs_2512_03715
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DINO-RotateMatch: A Rotation-Aware Deep Framework for Robust Image Matching in Large-Scale 3D Reconstruction
Zhang, Kaichen
Sheng, Tianxiang
Shi, Xuanming
Computer Vision and Pattern Recognition
This paper presents DINO-RotateMatch, a deep-learning framework designed to address the chal lenges of image matching in large-scale 3D reconstruction from unstructured Internet images. The method integrates a dataset-adaptive image pairing strategy with rotation-aware keypoint extraction and matching. DINO is employed to retrieve semantically relevant image pairs in large collections, while rotation-based augmentation captures orientation-dependent local features using ALIKED and Light Glue. Experiments on the Kaggle Image Matching Challenge 2025 demonstrate consistent improve ments in mean Average Accuracy (mAA), achieving a Silver Award (47th of 943 teams). The results confirm that combining self-supervised global descriptors with rotation-enhanced local matching offers a robust and scalable solution for large-scale 3D reconstruction.
title DINO-RotateMatch: A Rotation-Aware Deep Framework for Robust Image Matching in Large-Scale 3D Reconstruction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.03715