aiqbench analysis
WAPR introduces a zero-shot foundation model for refining 6D object poses in wide-angle scenarios, addressing the need for accuracy and speed with unseen objects. By leveraging rotational symmetry priors and a large-scale SA6D dataset, the method achieves state-of-the-art performance while maintaining high inference throughput. This work advances practical computer vision capabilities essential for robotics and augmented reality applications.
WAPR: A Foundation Model for Wide-Angle Refinement in Unseen Object Pose Estimation
arXiv:2610.09535v1 Announce Type: cross Abstract: Real-world applications require 6D pose estimation to be accurate, fast, and scalable to unseen objects. This paper introduces WAPR, a zero-shot wide-angle pose refinement model that refines candidate poses with rotational deviations up to 90 degrees. With as few as 12 candidate poses per detected object instance, WAPR supports fast inference within 1 s per frame and reaches a pose-estimation throughput of up to 25 detected object instances per second. To support wide-angle training for rotationally symmetric objects, WAPR uses rotational symmetry priors to canonicalize symmetry-equivalent pose targets before loss computation. We further construct SA6D, a large-scale 6D training dataset with such priors. SA6D obtains KASAL-assisted rotational symmetry priors for 944 GSO scans and expands them through geometry and texture augmentation into about 50K augmented object instances and about 2M rendered RGB-D images. In addition, an angle-balanced loss stabilizes learning across different angular ranges by reducing the influence of uninformative large-error cases. Experiments on seven BOP core datasets show that WAPR achieves state-of-the-art performance in unseen-object 6D pose localization and detection under both fast and unconstrained inference settings. Project page: https://github.com/WangYuLin-SEU/WAPR.
Abstract:Real-world applications require 6D pose estimation to be accurate, fast, and scalable to unseen objects. This paper introduces WAPR, a zero-shot wide-angle pose refinement model that refines candidate poses with rotational deviations up to 90 degrees. With as few as 12 candidate poses per detected object instance, WAPR supports fast inference within 1 s per frame and reaches a pose-estimation throughput of up to 25 detected object instances per second. To support wide-angle training for rotationally symmetric objects, WAPR uses rotational symmetry priors to canonicalize symmetry-equivalent pose targets before loss computation. We further construct SA6D, a large-scale 6D training dataset with such priors. SA6D obtains KASAL-assisted rotational symmetry priors for 944 GSO scans and expands them through geometry and texture augmentation into about 50K augmented object instances and about 2M rendered RGB-D images. In addition, an angle-balanced loss stabilizes learning across different angular ranges by reducing the influence of uninformative large-error cases. Experiments on seven BOP core datasets show that WAPR achieves state-of-the-art performance in unseen-object 6D pose localization and detection under both fast and unconstrained inference settings. Project page: this https URL.
Accepted to ECCV 2026. 19 pages, 4 figures. Yulin Wang and Mengting Hu contributed equally. Corresponding author: Chen Luo
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
arXiv:2610.09535 [cs.CV]
(or arXiv:2610.09535v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2610.09535 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
In: Computer Vision - ECCV 2026, LNCS vol. 17016, pp. 221-239. Springer, Cham (2026)
https://doi.org/10.1007/978-3-032-37383-0_13 Focus to learn more DOI(s) linking to related resources
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.
Full text under the publisher's open licence · 328 words
Related coverage
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · Sep 17, 2026
NVIDIA Omniverse NuRec adapts perception to new vehicle designs
Quantum Zeitgeist · Sep 8, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
Plan Along the Way: Event-Triggered Foundation-Model Planning for TAMP Execution in Partially Observable Manipulation
arXiv · Aug 31, 2026