Abstract:Visual localization refers to the process of determining camera poses and orientation within a known scene representation. This task is often complicated by factors such as illumination changes and variations in viewing angles. In this paper, we propose HGSLoc, a novel lightweight, plug and-play pose optimization framework, which integrates 3D reconstruction with a heuristic refinement strategy to achieve higher pose estimation accuracy. Specifically, we introduce an explicit geometric map for 3D representation and high-fidelity rendering, allowing the generation of high-quality synthesized views to support accurate visual localization. Our method demonstrates a faster rendering speed and higher localization accuracy compared to NeRF-based neural rendering localization approaches. We introduce a heuristic refinement strategy, its efficient optimization capability can quickly locate the target node, while we set the step-level optimization step to enhance the pose accuracy in the scenarios with small errors. With carefully designed heuristic functions, it offers efficient optimization capabilities, enabling rapid error reduction in rough localization estimations. Our method mitigates the dependence on complex neural network models while demonstrating improved robustness against noise and higher localization accuracy in challenging environments, as compared to neural network joint optimization strategies. The optimization framework proposed in this paper introduces novel approaches to visual localization by integrating the advantages of 3D reconstruction and heuristic refinement strategy, which demonstrates strong performance across multiple benchmark datasets, including 7Scenes and DB dataset.

Vision-Based Pose Optimization Using Learned Metrics

Parameter-adaptive Multi-Frame Joint Pose Optimization Method

Learning Stereopsis from Geometric Synthesis for 6D Object Pose Estimation

An Optimization Based Vision Calibration Method for PTZ Camera's Errors in Model and Execution

A SLAM Pose Graph Optimization Method Using Dual Visual Odometry

Image Gradient-based Joint Direct Visual Odometry for Stereo Camera

An Optimization Based Framework for Human Pose Estimation

Optimization Design of Camera Pose for Network Base on Multi-view Stereo Vision

Pose Refinement: Bridging the Gap Between Unsupervised Learning and Geometric Methods for Visual Odometry.

WPO-Net: Windowed Pose Optimization Network for Monocular Visual Odometry Estimation

Self-Supervised Deep Visual Odometry Based on Geometric Attention Model

Improving Robustness for Joint Optimization of Camera Poses and Decomposed Low-Rank Tensorial Radiance Fields

Nothing But Geometric Constraints: A Model-Free Method for Articulated Object Pose Estimation

RegNet: Learning the Optimization of Direct Image-to-Image Pose Registration

Geometric Loss Functions for Camera Pose Regression with Deep Learning

Robust self-supervised monocular visual odometry based on prediction-update pose estimation network.

HGSLoc: 3DGS-based Heuristic Camera Pose Refinement

A Hybrid Approach for Cross-modality Pose Estimation Between Image and Point Cloud

Hybrid bundle-adjusting 3D Gaussians for view consistent rendering with pose optimization

Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose

SD-Pose: facilitating space-decoupled human pose estimation via adaptive pose perception guidance