analiza aiqbench
This research introduces Con-DSO, a novel framework for RGB-D direct sparse odometry that enhances robustness against dynamic objects and unreliable depth through learned consistency priors. By integrating a dual-branch uncertainty model with decoupled photometric-geometric weighting, the method significantly reduces trajectory error across multiple benchmarks without relying on external sensors. The work advances the state of visual navigation by making depth-based pose estimation more resilient in complex, real-world environments.
Con-DSO: Learning Short-Horizon Consistency Priors for RGB-D Direct Sparse Odometry
arXiv:2605.27952v2 Announce Type: replace-cross Abstract: RGB-D direct visual odometry (VO) benefits from metric depth measurements but often degrades in the presence of dynamic objects, occlusions, illumination changes, and unreliable depth, which violate the photometric and geometric consistency assumptions of direct alignment. We propose Con-DSO, a consistency-aware RGB-D direct sparse odometry framework that addresses these challenges through a unified learned uncertainty model. A dual-branch consistency network is trained on adjacent RGB-D frame pairs using flow-guided photometric errors and projective depth-consistency errors to predict pixel-level photometric and geometric uncertainty. The predicted uncertainty is first converted into pairwise quality to guide support-pixel selection and is then fused across adjacent frame pairs to form a host-side quality prior for keyframe-based tracking. To account for the different roles of photometric and depth information in direct RGB-D optimization, the quality prior is incorporated through a decoupled photometric-geometric weighting scheme, with the geometric weight applied only to the translational component of pose estimation. Experiments on five public RGB-D benchmarks demonstrate consistent improvements over direct RGB-D odometry baselines, achieving more than 20\% reduction in absolute trajectory error on ICL-NUIM and approximately 50\% to 80\% reductions on RGB-D Scenes V2, TUM/BONN, and OpenLORIS. These results demonstrate that learned consistency-aware uncertainty can substantially improve the robustness of RGB-D direct visual odometry in challenging environments.
Abstract:RGB-D direct visual odometry (VO) benefits from metric depth measurements but often degrades in the presence of dynamic objects, occlusions, illumination changes, and unreliable depth, which violate the photometric and geometric consistency assumptions of direct alignment. We propose Con-DSO, a consistency-aware RGB-D direct sparse odometry framework that addresses these challenges through a unified learned uncertainty model. A dual-branch consistency network is trained on adjacent RGB-D frame pairs using flow-guided photometric errors and projective depth-consistency errors to predict pixel-level photometric and geometric uncertainty. The predicted uncertainty is first converted into pairwise quality to guide support-pixel selection and is then fused across adjacent frame pairs to form a host-side quality prior for keyframe-based tracking. To account for the different roles of photometric and depth information in direct RGB-D optimization, the quality prior is incorporated through a decoupled photometric-geometric weighting scheme, with the geometric weight applied only to the translational component of pose estimation. Experiments on five public RGB-D benchmarks demonstrate consistent improvements over direct RGB-D odometry baselines, achieving more than 20\% reduction in absolute trajectory error on ICL-NUIM and approximately 50\% to 80\% reductions on RGB-D Scenes V2, TUM/BONN, and OpenLORIS. These results demonstrate that learned consistency-aware uncertainty can substantially improve the robustness of RGB-D direct visual odometry in challenging environments.
Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
arXiv:2605.27952 [cs.CV]
(or arXiv:2605.27952v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2605.27952 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 306 cuvinte
Subiecte conexe
[AINews] Opus 5.5 is good at explainer videos
Latent Space · 29 sept. 2026
[AINews] The Future of Latent Space
Latent Space · 25 sept. 2026
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · 12 iun. 2026
Informe de riesgos de la IA de frontera (febrero–marzo de 2026)
METR · 19 mai 2026