arXiv subjects

Huaiyuan Weng

Publications and source records attributed to Huaiyuan Weng.

4 recordsLinked to original sources

arXiv2026

GS-CPE: Unified 6-Degree-of-Freedom Camera Pose Estimation via 3D Gaussian Splatting

Despite substantial progress in visual localization, from scene coordinate regression to direct camera pose regression, achieving both robust generalization and high accuracy remain challenging. This study introduces GS-CPE (Gaussian Splatting based Camera Pose Estimation), a coarse-to-fine framework for 6-DoF camera pose estimation that unifies geometry-based coarse pose estimation with robust 3D Gaussian Splatting (3DGS) warping based pose refinement. GS-CPE first estimates a coarse pose via retrieval-guided geometric pose estimation on a 3DGS scene representation, then refines it by minimizing a visibility aware masked RGB warping objective in a multi-scale optimization framework, with adaptive re-rendering. Extensive experiments on indoor and outdoor benchmarks including 7Scenes, Cambridge Landmarks, FAST-LIVO2 datasets, and a custom dataset demonstrate state-of-the-art performance, consistently outperforming in both accuracy and generalization.

cs.CV↗

arXiv2026

Structured-Li-GS: Structured 3D Gaussians Splatting with LiDAR Incorporation and Spatial Constraints

In this study, we develop a Structured framework for Gaussian Splatting (3DGS) with LiDAR integration (Structured-Li-GS). It is a lightweight Gaussian Splatting pipeline that leverages LiDAR-inertial-visual SLAM. Structured-Li-GS achieves high-quality 3D reconstructions with fewer Gaussians by training on accurate, dense, colorized point clouds. Gaussian primitives are anchored using sub-sampled point clouds, and their ellipsoidal parameters are initialized from local surface geometry. Our training strategy integrates a comprehensive set of loss terms, including photometric, flattening, offset, depth, and normal losses, guided by the dense point cloud, enabling accurate reconstruction without Gaussian densification. This approach produces up-to-scale, high-fidelity results with a moderate model size. For experimental validation, we develop a custom hardware-synchronized LiDAR-camera handheld scanner. Experiments on both benchmark datasets and our real-world in-house dataset demonstrate that Structured-Li-GS surpasses state-of-the-art methods while using fewer Gaussians.

cs.CV↗

arXiv2026

View Invariant Learning for Vision-Language Navigation in Continuous Environments

Vision-Language Navigation in Continuous Environments (VLNCE), where an agent follows instructions and moves freely to reach a destination, is a key research problem in embodied AI. However, most existing approaches are sensitive to viewpoint changes, i.e. variations in camera height and viewing angle. Here we introduce a more general scenario, V$^2$-VLNCE (VLNCE with Varied Viewpoints) and propose a view-invariant post-training framework, called VIL (View Invariant Learning), that makes existing navigation policies more robust to changes in camera viewpoint. VIL employs a contrastive learning framework to learn sparse and view-invariant features. We also introduce a teacher-student framework for the Waypoint Predictor Module, a standard part of VLNCE baselines, where a view-dependent teacher model distills knowledge into a view-invariant student model. We employ an end-to-end training paradigm to jointly optimize these components. Empirical results show that our method outperforms state-of-the-art approaches on V$^2$-VLNCE by 8-15\% measured on Success Rate for two standard benchmark datasets R2R-CE and RxR-CE. Evaluation of VIL in standard VLNCE settings shows that despite being trained for varied viewpoints, VIL often still improves performance. On the harder RxR-CE dataset, our method also achieved state-of-the-art performance across all metrics. This suggests that adding VIL does not diminish the standard viewpoint performance and can serve as a plug-and-play post-training method. We further evaluate VIL for simulated camera placements derived from real robot configurations (e.g. Stretch RE-1, LoCoBot), showing consistent improvements of performance. Finally, we present a proof-of-concept real-robot evaluation in two physical environments using a panoramic RGB sensor combined with LiDAR. The code is available at https://github.com/realjoshqsun/V2-VLNCE.

cs.CV↗

arXiv2025

A Target-Based Extrinsic Calibration Framework for Non-Overlapping Camera-Lidar Systems Using a Motion Capture System

We present a novel target-based lidar-camera extrinsic calibration methodology that can be used for non-overlapping field of view (FOV) sensors. Contrary to previous work, our methodology overcomes the non-overlapping FOV challenge using a motion capture system (MCS) instead of traditional simultaneous localization and mapping approaches. Due to the high relative precision of MCSs, our methodology can achieve both the high accuracy and repeatable calibrations common to traditional target-based methods, regardless of the amount of overlap in the sensors' field of view. Furthermore, we design a target-agnostic implementation that does not require uniquely identifiable features by using an iterative closest point approach, enabled by the MSC measurements. We show using simulation that we can accurately recover extrinsic calibrations for a range of perturbations to the true calibration that would be expected in real circumstances. We prove experimentally that our method out-performs state-of-the-art lidar-camera extrinsic calibration methods that can be used for non-overlapping FOV systems, while using a target-based approach that guarantees repeatably high accuracy. Lastly, we show in simulation that different target designs can be used, including easily constructed 3D targets such as a cylinder that are normally considered degenerate in most calibration formulations.

cs.RO↗