FoundationGeo: Learning Spatial Pixel-Wise Fields for Monocular Metric Geometry

• URL: http://arxiv.org/abs/2607.11588v3
• Date: Wed, 15 Jul 2026 08:02:14 GMT
• Title: FoundationGeo: Learning Spatial Pixel-Wise Fields for Monocular Metric Geometry
• Abstract summary: FoundationGeo bridges relative and metric prediction via spatial calibration and principled data design.<n>Stage 1 learns a high-fidelity, affine-invariant geometry model by initializing with DINOv3.<n>Stage 2 moves beyond global scaling by introducing pixel-wise calibration fields for metric estimation.
• Score: 52.232826445242644
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present FoundationGeo, a two-stage framework that explicitly bridges relative and metric prediction via spatial calibration and principled data design. Stage 1 learns a high-fidelity, affine-invariant geometry model by initializing with DINOv3 and training on a curated 10.2M-sample multi-domain corpus with complementary local-detail supervision, yielding sharp boundaries and strong cross-domain generalization. Stage 2 moves beyond global scaling by introducing lightweight pixel-wise calibration fields for metric estimation: a scale field for spatially varying metric alignment and a ray-direction correction field that mitigates directional bias in point-map geometry, together producing metrically consistent 3D point maps. Beyond model design, we identify camera intrinsic coverage, especially focal length distribution mismatch between training and test data, as a key bottleneck for zero-shot metric generalization: performance drops sharply when test intrinsics fall outside the training distribution. To address this, we synthesize additional training data across diverse focal lengths using a Blender-based data engine, repairing under-covered focal regimes and improving robustness under intrinsic shift. Extensive zero-shot evaluations across seven benchmarks show that FoundationGeo significantly strengthens cross-domain robustness, staying near the top across diverse domains while avoiding the sharp cross-domain performance drops observed in other methods. This consistency translates into the best overall performance, surpassing heavier baselines by over 5.2% on average.

Related papers

• Multimodal-Multiresolution Foundation Model for Lunar Remote Sensing [5.471205088925021]
  We present a multimodal foundation model for lunar remote sensing, pretrained from scratch on SomBench.<n>The model adapts the TerraMind masked-token architecture with two lunar-specific extensions.<n>We evaluate on four benchmarks: crater detection at WAC and NAC scales, irregular mare patch (IMP) segmentation, and polar ice prospectivity regression.
  arXiv  Detail & Related papers  (2026-09-08T14:55:35Z)
• UnderOneFacade: Worldwide Facade Semantic Segmentation Benchmark Dataset [19.044731989703468]
  We introduce UnderOneFacade, the largest cross-country and cross-continent 3D facade benchmark to date.<n>We show that current methods struggle to recognize fine-grained architectural elements and degrade significantly across geographic domains.<n>UnderOneFacade establishes a rigorous benchmark for developing robust and transferable 3D segmentation models.
  arXiv  Detail & Related papers  (2026-07-02T10:50:01Z)
• Efficient Hybrid CNN-GNN Architecture for Monocular Depth Estimation [0.0]
  We present GraphDepth, a monocular depth estimation architecture that integrates Graph Neural Networks (GNNs) within a convolutional encoder-decoder framework.<n>Our approach embeds efficient GraphSAGE layers at multiple scales of a ResNet-101 U-Net backbone.<n> Experiments on NYU Depth V2, WHU Aerial, ETH3D, and Mid-Air benchmarks demonstrate competitive accuracy within 4.6% of state-of-the-art transformers.
  arXiv  Detail & Related papers  (2026-05-11T09:21:04Z)
• XD-MAP: Cross-Modal Domain Adaptation using Semantic Parametric Mapping [5.609281438287908]
  We propose a novel approach to transferring sensor-specific knowledge from an image dataset to LiDAR.<n>Our method XD-MAP leverages detections from a neural network on camera images to create a semantic parametric map.<n>The results demonstrate the effectiveness of our approach achieving strong performance on LiDAR data without any manual labeling.
  arXiv  Detail & Related papers  (2026-01-20T21:00:26Z)
• 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)
• Topology-Aware Modeling for Unsupervised Simulation-to-Reality Point Cloud Recognition [63.55828203989405]
  We introduce a novel Topology-Aware Modeling (TAM) framework for Sim2Real UDA on object point clouds.<n>Our approach mitigates the domain gap by leveraging global spatial topology, characterized by low-level, high-frequency 3D structures.<n>We propose an advanced self-training strategy that combines cross-domain contrastive learning with self-training.
  arXiv  Detail & Related papers  (2025-06-26T11:53:59Z)
• TraIL-Det: Transformation-Invariant Local Feature Networks for 3D LiDAR Object Detection with Unsupervised Pre-Training [21.56675189346088]
  We introduce Transformation-Invariant Local (TraIL) features and the associated TraIL-Det architecture. TraIL features exhibit rigid transformation invariance and effectively adapt to variations in point density. They utilize the inherent isotropic radiation of LiDAR to enhance local representation. Our method outperforms contemporary self-supervised 3D object detection approaches in terms of mAP on KITTI.
  arXiv  Detail & Related papers  (2024-08-25T17:59:17Z)
• Boosting Cross-Domain Point Classification via Distilling Relational Priors from 2D Transformers [59.0181939916084]
  Traditional 3D networks mainly focus on local geometric details and ignore the topological structure between local geometries. We propose a novel Priors Distillation (RPD) method to extract priors from the well-trained transformers on massive images. Experiments on the PointDA-10 and the Sim-to-Real datasets verify that the proposed method consistently achieves the state-of-the-art performance of UDA for point cloud classification.
  arXiv  Detail & Related papers  (2024-07-26T06:29:09Z)
• Unleash the Potential of 3D Point Cloud Modeling with A Calibrated Local Geometry-driven Distance Metric [62.365983810610985]
  We propose a novel distance metric called Calibrated Local Geometry Distance (CLGD) CLGD computes the difference between the underlying 3D surfaces calibrated and induced by a set of reference points. As a generic metric, CLGD has the potential to advance 3D point cloud modeling.
  arXiv  Detail & Related papers  (2023-06-01T11:16:20Z)
• CL3D: Unsupervised Domain Adaptation for Cross-LiDAR 3D Detection [16.021932740447966]
  Domain adaptation for Cross-LiDAR 3D detection is challenging due to the large gap on the raw data representation. We present an unsupervised domain adaptation method that overcomes above difficulties.
  arXiv  Detail & Related papers  (2022-12-01T03:22:55Z)
• Pretrained equivariant features improve unsupervised landmark discovery [69.02115180674885]
  We formulate a two-step unsupervised approach that overcomes this challenge by first learning powerful pixel-based features. Our method produces state-of-the-art results in several challenging landmark detection datasets.
  arXiv  Detail & Related papers  (2021-04-07T05:42:11Z)

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.