gravity direction estimation

This work addresses the limitations of traditional 3D reconstruction methods, which predict point maps in camera-centered coordinates, struggle to incorporate scene structural priors, and suffer from high rotational degrees of freedom across views, leading to inconsistent reconstructions. To overcome these issues, the authors propose predicting point maps in a gravity-aligned upright coordinate system, thereby reducing inter-view rotational ambiguity through a shared vertical axis. They introduce the Gravity Grounded Geometry Transformer (G3T) model and the G3T-Long incremental reconstruction framework, which for the first time integrate gravity-aligned coordinates into point map prediction by combining a Transformer architecture, gravity-aware pose estimation, and a submap stitching strategy. Experiments demonstrate that this approach significantly improves reconstruction accuracy and robustness, outperforming existing methods in incremental 3D reconstruction and validating the effectiveness of gravity-aligned representations.

This work addresses the susceptibility of IMU-provided gravity priors to interference from linear acceleration or vibration, a challenge that existing methods struggle to mitigate using only a single image. To this end, we propose GravCal, an end-to-end feedforward network that fuses image-based independent gravity estimation with a residual correction pathway. A learnable gating mechanism adaptively weights these components, jointly outputting a refined gravity direction and a confidence score correlated with prior quality. GravCal is the first method capable of correcting noisy IMU gravity priors using only a single RGB image. Evaluated on a newly curated dataset of 148K frames, it reduces the mean angular error from 22.02° to 14.24°, with particularly pronounced improvements under severely degraded priors. The predicted confidence effectively guides downstream tasks.

This paper addresses the problem of multi-view relative pose estimation under known vertical direction (e.g., provided by an IMU). To reduce correspondence requirements, it leverages vertical prior knowledge to constrain the pose space to two rotational degrees of freedom and two translational components. The method introduces the first minimal (three-point) and linear closed-form (four-point) solutions for trifocal pose estimation. It integrates trifocal tensor geometry, Gröbner basis algebraic solving, and linear least-squares refinement—achieving high accuracy while significantly improving RANSAC robustness and computational efficiency. Experiments on the KITTI and synthetic datasets demonstrate that the proposed approach outperforms state-of-the-art methods in pose estimation accuracy.

This work addresses the limitation of existing RGB-based multi-view reconstruction methods, which produce monolithic scene representations lacking explicit physical structure and thus hinder stable physical interaction. The authors propose an end-to-end reconstruction framework that relies solely on RGB images and introduces gravity as a universal physical prior. By aligning views in a gravity-consistent coordinate system, reconstructing object-centric rigid-body meshes, and employing conditional 3D point classification to remove background redundancy, the method decouples foreground objects from background geometry without requiring CAD model retrieval. The output is a structured hybrid representation suitable for simulation. Experiments demonstrate significant improvements over retrieval-based baselines in 6-DoF object pose accuracy, decoupling quality, and rendering-to-simulation efficiency, both in simulated and real-world scenes.

Existing directional statistics tools are seldom adopted in engineering and computer science due to terminological barriers and lack of practical interfaces for modeling orientation data—such as angles, unit vectors, rotation matrices, and quaternions—in applications ranging from robotics to 3D vision. Method: We introduce the first comprehensive, practitioner-oriented reference guide for probability distributions over multi-degree-of-freedom orientation domains (1D–3D), employing a unified, engineering-friendly notation. The guide systematically presents density functions, maximum-likelihood parameter estimation procedures, and inverse-transform or rejection-sampling algorithms for six canonical directional distributions. Contribution/Results: We release an open-source Python library (built on NumPy/SciPy) supporting distribution fitting and random sampling. Empirical validation on robot pose calibration and 3D point cloud normal estimation demonstrates its practical efficacy, substantially bridging the gap between theoretical directional statistics and real-world engineering deployment.

During motion, inertial measurement units (IMUs) are subject to non-gravitational accelerations—such as centripetal and tangential components—which distort gravity direction estimation, particularly when the IMU is mounted far from the system’s center of rotation. This work proposes a self-calibration method that requires no external equipment and jointly estimates IMU intrinsic parameters (axis misalignment, bias, and scale factors) and extrinsic parameters (the displacement vector from the base frame to the sensor). Crucially, it explicitly models motion-induced acceleration using gyroscope data to compensate gravity observations within the attitude estimation pipeline. To the best of our knowledge, this is the first approach to integrate motion acceleration compensation directly into orientation estimation, enabling seamless compatibility with established filters such as Madgwick and Mahony. Experimental results demonstrate that the proposed framework significantly improves attitude accuracy under high-dynamic conditions and non-central IMU mounting configurations.