Salvato in:
Dettagli Bibliografici
Autori principali: Lam, Chit Yuen, Clark, Ronald, Kocer, Basaran Bahadir
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2503.00315
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910851802333184
author Lam, Chit Yuen
Clark, Ronald
Kocer, Basaran Bahadir
author_facet Lam, Chit Yuen
Clark, Ronald
Kocer, Basaran Bahadir
contents We introduce XIRVIO, a transformer-based Generative Adversarial Network (GAN) framework for monocular visual inertial odometry (VIO). By taking sequences of images and 6-DoF inertial measurements as inputs, XIRVIO's generator predicts pose trajectories through an iterative refinement process which are then evaluated by the critic to select the iteration with the optimised prediction. Additionally, the self-emergent adaptive sensor weighting reveals how XIRVIO attends to each sensory input based on contextual cues in the data, making it a promising approach for achieving explainability in safety-critical VIO applications. Evaluations on the KITTI dataset demonstrate that XIRVIO matches well-known state-of-the-art learning-based methods in terms of both translation and rotation errors.
format Preprint
id arxiv_https___arxiv_org_abs_2503_00315
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle XIRVIO: Critic-guided Iterative Refinement for Visual-Inertial Odometry with Explainable Adaptive Weighting
Lam, Chit Yuen
Clark, Ronald
Kocer, Basaran Bahadir
Robotics
We introduce XIRVIO, a transformer-based Generative Adversarial Network (GAN) framework for monocular visual inertial odometry (VIO). By taking sequences of images and 6-DoF inertial measurements as inputs, XIRVIO's generator predicts pose trajectories through an iterative refinement process which are then evaluated by the critic to select the iteration with the optimised prediction. Additionally, the self-emergent adaptive sensor weighting reveals how XIRVIO attends to each sensory input based on contextual cues in the data, making it a promising approach for achieving explainability in safety-critical VIO applications. Evaluations on the KITTI dataset demonstrate that XIRVIO matches well-known state-of-the-art learning-based methods in terms of both translation and rotation errors.
title XIRVIO: Critic-guided Iterative Refinement for Visual-Inertial Odometry with Explainable Adaptive Weighting
topic Robotics
url https://arxiv.org/abs/2503.00315