GeoMix: Descriptor-Free Visual Localization via Global Context and Multi-Detector Training

• URL: http://arxiv.org/abs/2607.02486v1
• Date: Thu, 02 Jul 2026 17:52:41 GMT
• Title: GeoMix: Descriptor-Free Visual Localization via Global Context and Multi-Detector Training
• Abstract summary: Descriptor-free visual localization eliminates high-dimensional descriptor storage, preserves scene privacy, and simplifies map maintenance.<n>We identify this gap to insufficient geometric discriminability in geometry-only matching.<n>We propose GeoMix, a descriptor-free 2D-3D matching framework that strengthens geometric discriminability at three levels.
• Score: 27.71531304367363
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Descriptor-free visual localization eliminates high-dimensional descriptor storage, preserves scene privacy, and simplifies map maintenance, yet its accuracy still lags far behind descriptor-based pipelines. We identify this gap to insufficient geometric discriminability in geometry-only matching. Without visual appearance, current methods underutilize local geometry cues, lack the global context among keypoints, and overfit to a single keypoint detector. We further observe that descriptor-free matching naturally enables multi-detector training, as heterogeneous keypoints can be optimized in a shared geometry-only space without aligning descriptor spaces. Building on these insights, we propose GeoMix, a descriptor-free 2D-3D matching framework that strengthens geometric discriminability at three levels. Locally, directional and distance-aware embeddings enrich neighborhood aggregation with fine-grained spatial structure. Globally, learnable context nodes aggregate and redistribute scene-wide information via cross-attention to resolve ambiguities beyond local receptive fields. At the training level, Mix-Training exploits this detector-agnostic geometry space to learn representations across multiple keypoint detectors. Extensive experiments on MegaDepth, Cambridge Landmarks, 7Scenes, and Aachen Day-Night show that GeoMix sets a new state of the art among descriptor-free methods, reducing 75th-percentile rotation error by 89\% and translation error by up to 90\% over the previous best, while generalizing zero-shot to unseen detectors and narrowing the gap to descriptor-based pipelines. Code is available at $\href{https://github.com/YejunZhang/Geomix}{\text{this links}}$.

Related papers

• Robust Scene Coordinate Regression via Geometrically-Consistent Global Descriptors [52.57327385675752]
  We propose an aggregator module that learns global descriptors consistent with both geometrical structure and visual similarity.<n>This corrects erroneous associations caused by unreliable overlap scores.<n>Experiments on challenging benchmarks show substantial localization gains in large-scale environments.
  arXiv  Detail & Related papers  (2025-12-19T04:24:03Z)
• NeuraLoc: Visual Localization in Neural Implicit Map with Dual Complementary Features [50.212836834889146]
  We propose an efficient and novel visual localization approach based on the neural implicit map with complementary features.<n>Specifically, to enforce geometric constraints and reduce storage requirements, we implicitly learn a 3D keypoint descriptor field.<n>To further address the semantic ambiguity of descriptors, we introduce additional semantic contextual feature fields.
  arXiv  Detail & Related papers  (2025-03-08T08:04:27Z)
• FUSELOC: Fusing Global and Local Descriptors to Disambiguate 2D-3D Matching in Visual Localization [52.57327385675752]
  Direct 2D-3D matching requires significantly less memory but suffers from lower accuracy due to the larger and more ambiguous search space.<n>We address this ambiguity by fusing local and global descriptors using a weighted average operator.<n>We achieve performance close to hierarchical methods while using 43% less memory and running 1.6 times faster.
  arXiv  Detail & Related papers  (2024-08-21T23:42:16Z)
• GeoFormer: Learning Point Cloud Completion with Tri-Plane Integrated Transformer [41.26276375114911]
  Point cloud completion aims to recover accurate global geometry and preserve fine-grained local details from partial point clouds. Conventional methods typically predict unseen points directly from 3D point cloud coordinates or use self-projected multi-view depth maps. We introduce a GeoFormer that simultaneously enhances the global geometric structure of the points and improves the local details.
  arXiv  Detail & Related papers  (2024-08-13T03:15:36Z)
• Coupled Laplacian Eigenmaps for Locally-Aware 3D Rigid Point Cloud Matching [0.0]
  We propose a new technique, based on graph Laplacian eigenmaps, to match point clouds by taking into account fine local structures. To deal with the order and sign ambiguity of Laplacian eigenmaps, we introduce a new operator, called Coupled Laplacian. We show that the similarity between those aligned high-dimensional spaces provides a locally meaningful score to match shapes.
  arXiv  Detail & Related papers  (2024-02-27T10:10:12Z)
• Uni3DETR: Unified 3D Detection Transformer [75.35012428550135]
  We propose a unified 3D detector that addresses indoor and outdoor detection within the same framework. Specifically, we employ the detection transformer with point-voxel interaction for object prediction. We then propose the mixture of query points, which sufficiently exploits global information for dense small-range indoor scenes and local information for large-range sparse outdoor ones.
  arXiv  Detail & Related papers  (2023-10-09T13:20:20Z)
• Flattening-Net: Deep Regular 2D Representation for 3D Point Cloud Analysis [66.49788145564004]
  We present an unsupervised deep neural architecture called Flattening-Net to represent irregular 3D point clouds of arbitrary geometry and topology. Our methods perform favorably against the current state-of-the-art competitors.
  arXiv  Detail & Related papers  (2022-12-17T15:05:25Z)
• RIGA: Rotation-Invariant and Globally-Aware Descriptors for Point Cloud Registration [44.23935553097983]
  We introduce RIGA to learn descriptors that are Rotation-Invariant by design and Globally-Aware. RIGA surpasses the state-of-the-art methods by a margin of 8degree in terms of the Relative Rotation Error on ModelNet40 and improves the Feature Matching Recall by at least 5 percentage points on 3DLoMatch.
  arXiv  Detail & Related papers  (2022-09-27T08:45:56Z)
• Viewpoint Invariant Dense Matching for Visual Geolocalization [15.8038460597256]
  We propose a novel method for image matching based on dense local features and tailored for visual geolocalization. Our method, called GeoWarp, directly embeds invariance to viewpoint shifts in the process of extracting dense features. GeoWarp is implemented efficiently as a re-ranking method that can be easily embedded into pre-existing visual geolocalization pipelines.
  arXiv  Detail & Related papers  (2021-09-20T20:17:38Z)
• Distinctive 3D local deep descriptors [2.512827436728378]
  Point cloud patches are extracted, canonicalised with respect to their estimated local reference frame and encoded by a PointNet-based deep neural network. We evaluate and compare DIPs against alternative hand-crafted and deep descriptors on several datasets consisting of point clouds reconstructed using different sensors.
  arXiv  Detail & Related papers  (2020-09-01T06:25:06Z)
• DH3D: Deep Hierarchical 3D Descriptors for Robust Large-Scale 6DoF Relocalization [56.15308829924527]
  We propose a Siamese network that jointly learns 3D local feature detection and description directly from raw 3D points. For detecting 3D keypoints we predict the discriminativeness of the local descriptors in an unsupervised manner. Experiments on various benchmarks demonstrate that our method achieves competitive results for both global point cloud retrieval and local point cloud registration.
  arXiv  Detail & Related papers  (2020-07-17T20:21:22Z)

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.