WAPR: A Foundation Model for Wide-Angle Refinement in Unseen Object Pose Estimation

Key points

• Real-world applications require 6D pose estimation to be accurate, fast, and scalable to unseen objects.1
• To support wide-angle training for rotationally symmetric objects, WAPR uses rotational symmetry priors to canonicalize symmetry-equivalent pose targets before loss computation.1
• 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.1
• 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.1

Sources (1)

• [1]WAPR: A Foundation Model for Wide-Angle Refinement in Unseen Object Pose Estimation
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 06:38 AM

  This paper introduces WAPR, a zero-shot wide-angle pose refinement model that refines candidate poses with rotational deviations up to 90 degrees.

  Real-world applications require 6D pose estimation to be accurate, fast, and scalable to unseen objects.

Extractive summary: sentences quoted from the sources.