Self-Supervised Cross-Modal Learning for Image-to-Point Cloud Registration

• URL: http://arxiv.org/abs/2509.15882v1
• Date: Fri, 19 Sep 2025 11:29:22 GMT
• Title: Self-Supervised Cross-Modal Learning for Image-to-Point Cloud Registration
• Authors: Xingmei Wang, Xiaoyu Hu, Chengkai Huang, Ziyan Zeng, Guohao Nie, Quan Z. Sheng, Lina Yao,
• Abstract summary: CrossI2P is a self-supervised framework that unifies cross-modal learning and two-stage registration in a single end-to-end pipeline.<n>We show that CrossI2P outperforms state-of-the-art methods by 23.7% on the KITTI Odometry benchmark and by 37.9% on nuScenes.
• Score: 22.360139236823155
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Bridging 2D and 3D sensor modalities is critical for robust perception in autonomous systems. However, image-to-point cloud (I2P) registration remains challenging due to the semantic-geometric gap between texture-rich but depth-ambiguous images and sparse yet metrically precise point clouds, as well as the tendency of existing methods to converge to local optima. To overcome these limitations, we introduce CrossI2P, a self-supervised framework that unifies cross-modal learning and two-stage registration in a single end-to-end pipeline. First, we learn a geometric-semantic fused embedding space via dual-path contrastive learning, enabling annotation-free, bidirectional alignment of 2D textures and 3D structures. Second, we adopt a coarse-to-fine registration paradigm: a global stage establishes superpoint-superpixel correspondences through joint intra-modal context and cross-modal interaction modeling, followed by a geometry-constrained point-level refinement for precise registration. Third, we employ a dynamic training mechanism with gradient normalization to balance losses for feature alignment, correspondence refinement, and pose estimation. Extensive experiments demonstrate that CrossI2P outperforms state-of-the-art methods by 23.7% on the KITTI Odometry benchmark and by 37.9% on nuScenes, significantly improving both accuracy and robustness.

Related papers

• Preoperative-to-intraoperative Liver Registration for Laparoscopic Surgery via Latent-Grounded Correspondence Constraints [51.7011449975586]
  Land-Reg is a deformable registration framework that learns latent-grounded 2D-3D landmark correspondences.<n>For rigid registration, Land-Reg embraces a Cross-modal Latent Alignment module.<n>An Uncertainty-enhanced Overlap Landmark Detector with similarity matching is proposed to robustly estimate explicit 2D-3D landmark correspondences.
  arXiv  Detail & Related papers  (2026-03-02T10:44:03Z)
• Where, Not What: Compelling Video LLMs to Learn Geometric Causality for 3D-Grounding [0.8883733362171032]
  We propose a novel training framework called What-Where Representation Re-Forming (W2R2) to tackle this issue.<n>Our approach fundamentally reshapes the model's internal space by designating 2D features as semantic beacons for "What" identification and 3D features as spatial anchors for "Where" localization.<n>Experiments conducted on ScanRefer and ScanQA demonstrate the effectiveness of W2R2, with significant gains in localization accuracy and robustness.
  arXiv  Detail & Related papers  (2025-10-19T22:40:18Z)
• Zero-shot Inexact CAD Model Alignment from a Single Image [53.37898107159792]
  A practical approach to infer 3D scene structure from a single image is to retrieve a closely matching 3D model from a database and align it with the object in the image.<n>Existing methods rely on supervised training with images and pose annotations, which limits them to a narrow set of object categories.<n>We propose a weakly supervised 9-DoF alignment method for inexact 3D models that requires no pose annotations and generalizes to unseen categories.
  arXiv  Detail & Related papers  (2025-07-04T04:46:59Z)
• Cross-Modal Geometric Hierarchy Fusion: An Implicit-Submap Driven Framework for Resilient 3D Place Recognition [9.411542547451193]
  We propose a novel framework that redefines 3D place recognition through density-agnostic geometric reasoning.<n>Specifically, we introduce an implicit 3D representation based on elastic points, which is immune to the interference of original scene point cloud density.<n>With the aid of these two types of information, we obtain descriptors that fuse geometric information from both bird's-eye view and 3D segment perspectives.
  arXiv  Detail & Related papers  (2025-06-17T07:04:07Z)
• Learning to Align and Refine: A Foundation-to-Diffusion Framework for Occlusion-Robust Two-Hand Reconstruction [50.952228546326516]
  Two-hand reconstruction from monocular images faces persistent challenges due to complex and dynamic hand postures.<n>Existing approaches struggle with such alignment issues, often resulting in misalignment and penetration artifacts.<n>We propose a dual-stage Foundation-to-Diffusion framework that precisely align 2D prior guidance from vision foundation models.
  arXiv  Detail & Related papers  (2025-03-22T14:42:27Z)
• CMR-Agent: Learning a Cross-Modal Agent for Iterative Image-to-Point Cloud Registration [2.400446821380503]
  Image-to-point cloud registration aims to determine the relative camera pose of an RGB image with respect to a point cloud. Most learning-based methods establish 2D-3D point correspondences in feature space without any feedback mechanism for iterative optimization. We propose to reformulate the registration procedure as an iterative Markov decision process, allowing for incremental adjustments to the camera pose.
  arXiv  Detail & Related papers  (2024-08-05T11:40:59Z)
• SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning [17.99904937160487]
  We introduce SCIPaD, a novel approach that incorporates spatial clues for unsupervised depth-pose joint learning. SCIPaD achieves a reduction of 22.2% in average translation error and 34.8% in average angular error for camera pose estimation task on the KITTI Odometry dataset.
  arXiv  Detail & Related papers  (2024-07-07T06:52:51Z)
• Dynamic 3D Point Cloud Sequences as 2D Videos [81.46246338686478]
  3D point cloud sequences serve as one of the most common and practical representation modalities of real-world environments. We propose a novel generic representation called textitStructured Point Cloud Videos (SPCVs) SPCVs re-organizes a point cloud sequence as a 2D video with spatial smoothness and temporal consistency, where the pixel values correspond to the 3D coordinates of points.
  arXiv  Detail & Related papers  (2024-03-02T08:18:57Z)
• Self-supervised Learning of LiDAR 3D Point Clouds via 2D-3D Neural Calibration [107.61458720202984]
  This paper introduces a novel self-supervised learning framework for enhancing 3D perception in autonomous driving scenes.<n>We propose the learnable transformation alignment to bridge the domain gap between image and point cloud data.<n>We establish dense 2D-3D correspondences to estimate the rigid pose.
  arXiv  Detail & Related papers  (2024-01-23T02:41:06Z)
• Self-Supervised Multi-Frame Monocular Scene Flow [61.588808225321735]
  We introduce a multi-frame monocular scene flow network based on self-supervised learning. We observe state-of-the-art accuracy among monocular scene flow methods based on self-supervised learning.
  arXiv  Detail & Related papers  (2021-05-05T17:49:55Z)

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.