# Global-Aware Edge Prioritization for Pose Graph Initialization

arXiv: https://arxiv.org/abs/2602.21963 · HTML: https://mlsanity.com/paper/2602.21963/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 54 references that are in the map.
- First posted on arXiv 25 Feb 2026.

## Builds on (most cited first, 23 of 23)

- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [SuperGlue: Learning Feature Matching with Graph Neural Networks](https://mlsanity.com/paper/1911.11763/) (2019)
- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [MegaDepth: Learning Single-View Depth Prediction from Internet Photos](https://mlsanity.com/paper/1804.00607/) (2018)
- [LightGlue: Local Feature Matching at Light Speed](https://mlsanity.com/paper/2306.13643/) (2023)
- [Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition](https://mlsanity.com/paper/2103.01486/) (2021)
- [Global Structure-from-Motion Revisited](https://mlsanity.com/paper/2407.20219/) (2024)
- [Stable View Synthesis](https://mlsanity.com/paper/2011.07233/) (2020)
- [AnyLoc: Towards Universal Visual Place Recognition](https://mlsanity.com/paper/2308.00688/) (2023)
- [Optimal Transport Aggregation for Visual Place Recognition](https://mlsanity.com/paper/2311.15937/) (2023)
- [MAGSAC: marginalizing sample consensus](https://mlsanity.com/paper/1803.07469/) (2018)
- [MegaLoc: One Retrieval to Place Them All](https://mlsanity.com/paper/2502.17237/) (2025)
- [Light3R-SfM: Towards Feed-forward Structure-from-Motion](https://mlsanity.com/paper/2501.14914/) (2025)
- [Visual Camera Re-Localization Using Graph Neural Networks and Relative Pose Supervision](https://mlsanity.com/paper/2104.02538/) (2021)
- [Doppelgangers: Learning to Disambiguate Images of Similar Structures](https://mlsanity.com/paper/2309.02420/) (2023)
- [Context-Aware Learning to Rank with Self-Attention](https://mlsanity.com/paper/2005.10084/) (2020)
- [Graph-Based Parallel Large Scale Structure from Motion](https://mlsanity.com/paper/1912.10659/) (2019)
- [Visual Localization using Imperfect 3D Models from the Internet](https://mlsanity.com/paper/2304.05947/) (2023)
- [Doppelgangers++: Improved Visual Disambiguation with Geometric 3D Features](https://mlsanity.com/paper/2412.05826/) (2024)
- [Efficient Initial Pose-graph Generation for Global SfM](https://mlsanity.com/paper/2011.11986/) (2020)
- [Learning Structure-from-Motion with Graph Attention Networks](https://mlsanity.com/paper/2308.15984/) (2023)
- [Breaking the Frame: Visual Place Recognition by Overlap Prediction](https://mlsanity.com/paper/2406.16204/) (2024)

## Built on by (most cited first, 1 of 1)

- [GeoWeaver: Accurate Long-Sequence 3D Reconstruction via Hierarchical Geometric Assembly](https://mlsanity.com/paper/2608.17389/) (2026)

## Abstract (arXiv)

The pose graph is a core component of Structure-from-Motion (SfM), where images act as nodes and edges encode relative poses. Since geometric verification is expensive, SfM pipelines restrict the pose graph to a sparse set of candidate edges, making initialization critical. Existing methods rely on image retrieval to connect each image to its k nearest neighbors, treating pairs independently and ignoring global consistency. We address this limitation through the concept of edge prioritization, ranking candidate edges by their utility for SfM. Our approach has three components: (1) a GNN trained with SfM-derived supervision to predict globally consistent edge reliability; (2) multi-minimal-spanning-tree-based pose graph construction guided by these ranks; and (3) connectivity-aware score modulation that reinforces weak regions and reduces graph diameter. This globally informed initialization yields more reliable and compact pose graphs, improving reconstruction accuracy in sparse and high-speed settings and outperforming SOTA retrieval methods on ambiguous scenes. The ode and trained models are available at https://github.com/weitong8591/globaledgeprior.