DINOReg: Strong Point Cloud Registration with Vision Foundation Model

• URL: http://arxiv.org/abs/2509.24370v1
• Date: Mon, 29 Sep 2025 07:15:47 GMT
• Title: DINOReg: Strong Point Cloud Registration with Vision Foundation Model
• Authors: Congjia Chen, Yufu Qu,
• Abstract summary: Point cloud registration is a fundamental task in 3D computer vision.<n>Recent studies have incorporated color information from RGB-D data into feature extraction.<n>We propose DINOReg, a registration network that sufficiently utilizes both visual and geometric information.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Point cloud registration is a fundamental task in 3D computer vision. Most existing methods rely solely on geometric information for feature extraction and matching. Recently, several studies have incorporated color information from RGB-D data into feature extraction. Although these methods achieve remarkable improvements, they have not fully exploited the abundant texture and semantic information in images, and the feature fusion is performed in an image-lossy manner, which limit their performance. In this paper, we propose DINOReg, a registration network that sufficiently utilizes both visual and geometric information to solve the point cloud registration problem. Inspired by advances in vision foundation models, we employ DINOv2 to extract informative visual features from images, and fuse visual and geometric features at the patch level. This design effectively combines the rich texture and global semantic information extracted by DINOv2 with the detailed geometric structure information captured by the geometric backbone. Additionally, a mixed positional embedding is proposed to encode positional information from both image space and point cloud space, which enhances the model's ability to perceive spatial relationships between patches. Extensive experiments on the RGBD-3DMatch and RGBD-3DLoMatch datasets demonstrate that our method achieves significant improvements over state-of-the-art geometry-only and multi-modal registration methods, with a 14.2% increase in patch inlier ratio and a 15.7% increase in registration recall. The code is publicly available at https://github.com/ccjccjccj/DINOReg.

Related papers

• ViGG: Robust RGB-D Point Cloud Registration using Visual-Geometric Mutual Guidance [18.052751061895215]
  ViGG is a robust RGB-D registration method using mutual guidance.<n>Experiments on 3DMatch, ScanNet and KITTI datasets show that our method outperforms recent state-of-the-art methods in both learning-free and learning-based settings.
  arXiv  Detail & Related papers  (2025-11-28T06:27:37Z)
• Cross-modal feature fusion for robust point cloud registration with ambiguous geometry [6.742883954812066]
  We propose a novel Cross-modal Feature Fusion method for point cloud registration.<n>It incorporates a two-stage fusion of 3D point cloud features and 2D image features.<n>It achieves state-of-the-art registration performance across all benchmarks.
  arXiv  Detail & Related papers  (2025-05-19T13:22:46Z)
• DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation [66.7989548848166]
  Existing approaches encode depth maps along with RGB images and perform feature fusion between them to enable more robust predictions.<n>We present DFormerv2, a strong RGBD encoder that explicitly uses depth maps as geometry priors rather than encoding depth information with neural networks.<n>Our goal is to extract the geometry clues from the depth and spatial distances among all the image patch tokens, which will then be used as geometry priors to allocate attention weights in self-attention.
  arXiv  Detail & Related papers  (2025-04-07T03:06:07Z)
• ZeroReg: Zero-Shot Point Cloud Registration with Foundation Models [77.84408427496025]
  State-of-the-art 3D point cloud registration methods rely on labeled 3D datasets for training.<n>We introduce ZeroReg, a zero-shot registration approach that utilizes 2D foundation models to predict 3D correspondences.
  arXiv  Detail & Related papers  (2023-12-05T11:33:16Z)
• PointMBF: A Multi-scale Bidirectional Fusion Network for Unsupervised RGB-D Point Cloud Registration [6.030097207369754]
  We propose a network implementing multi-scale bidirectional fusion between RGB images and point clouds generated from depth images. Our method achieves new state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-08-09T08:13:46Z)
• GQE-Net: A Graph-based Quality Enhancement Network for Point Cloud Color Attribute [51.4803148196217]
  We propose a graph-based quality enhancement network (GQE-Net) to reduce color distortion in point clouds. GQE-Net uses geometry information as an auxiliary input and graph convolution blocks to extract local features efficiently. Experimental results show that our method achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-03-24T02:33:45Z)
• PCR-CG: Point Cloud Registration via Deep Explicit Color and Geometry [28.653015760036602]
  We introduce a novel 3D point cloud registration module explicitly embedding the color signals into the geometry representation. Our key contribution is a 2D-3D cross-modality learning algorithm that embeds the deep features learned from color signals to the geometry representation. Our study reveals a significant advantages of correlating explicit deep color features to the point cloud in the registration task.
  arXiv  Detail & Related papers  (2023-02-28T08:50:17Z)
• Improving RGB-D Point Cloud Registration by Learning Multi-scale Local Linear Transformation [38.64501645574878]
  Point cloud registration aims at estimating the geometric transformation between two point cloud scans. Recent point cloud registration methods have tried to apply RGB-D data to achieve more accurate correspondence. We propose a new Geometry-Aware Visual Feature Extractor (GAVE) that employs multi-scale local linear transformation.
  arXiv  Detail & Related papers  (2022-08-31T14:36:09Z)
• PointMCD: Boosting Deep Point Cloud Encoders via Multi-view Cross-modal Distillation for 3D Shape Recognition [55.38462937452363]
  We propose a unified multi-view cross-modal distillation architecture, including a pretrained deep image encoder as the teacher and a deep point encoder as the student. By pair-wise aligning multi-view visual and geometric descriptors, we can obtain more powerful deep point encoders without exhausting and complicated network modification.
  arXiv  Detail & Related papers  (2022-07-07T07:23:20Z)
• SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-Training for Spatial-Aware Visual Representations [85.38562724999898]
  We propose a 2D Image and 3D Point cloud Unsupervised pre-training strategy, called SimIPU. Specifically, we develop a multi-modal contrastive learning framework that consists of an intra-modal spatial perception module and an inter-modal feature interaction module. To the best of our knowledge, this is the first study to explore contrastive learning pre-training strategies for outdoor multi-modal datasets.
  arXiv  Detail & Related papers  (2021-12-09T03:27:00Z)
• RGB-D Saliency Detection via Cascaded Mutual Information Minimization [122.8879596830581]
  Existing RGB-D saliency detection models do not explicitly encourage RGB and depth to achieve effective multi-modal learning. We introduce a novel multi-stage cascaded learning framework via mutual information minimization to "explicitly" model the multi-modal information between RGB image and depth data.
  arXiv  Detail & Related papers  (2021-09-15T12:31:27Z)
• ImVoteNet: Boosting 3D Object Detection in Point Clouds with Image Votes [93.82668222075128]
  We propose a 3D detection architecture called ImVoteNet for RGB-D scenes. ImVoteNet is based on fusing 2D votes in images and 3D votes in point clouds. We validate our model on the challenging SUN RGB-D dataset.
  arXiv  Detail & Related papers  (2020-01-29T05:09:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.