DINO-RotateMatch: A Rotation-Aware Deep Framework for Robust Image Matching in Large-Scale 3D Reconstruction
By: Kaichen Zhang, Tianxiang Sheng, Xuanming Shi
Potential Business Impact:
Helps computers build 3D worlds from internet pictures.
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.
Similar Papers
RoMa v2: Harder Better Faster Denser Feature Matching
CV and Pattern Recognition
Makes computers see 3D scenes more accurately.
RoMa v2: Harder Better Faster Denser Feature Matching
CV and Pattern Recognition
Makes computer vision see better in tough scenes.
DINOv3
CV and Pattern Recognition
Teaches computers to see and understand images better.