RayPose: Ray Bundling Diffusion for Template Views in Unseen 6D Object Pose Estimation

2025-10-21

AI Summary

To address accuracy degradation in template-based 6D pose estimation caused by template retrieval failure, this paper proposes a ray-alignment paradigm: modeling pose estimation as a geometric alignment task between the query image and multi-view template images in camera-ray space. Key contributions include: (1) reparameterizing camera rays with the object center as origin to decouple and explicitly model rotation; (2) jointly constraining translation estimation via dense translation offsets and geometric priors (e.g., depth consistency); and (3) designing a coarse-to-fine diffusion Transformer architecture that achieves scale-invariant rotation prediction and dense displacement regression through a ray-bundle diffusion mechanism. The method achieves state-of-the-art performance across multiple benchmarks and demonstrates superior robustness and generalization to unseen objects.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Typical template-based object pose pipelines estimate the pose by retrieving the closest matching template and aligning it with the observed image. However, failure to retrieve the correct template often leads to inaccurate pose predictions. To address this, we reformulate template-based object pose estimation as a ray alignment problem, where the viewing directions from multiple posed template images are learned to align with a non-posed query image. Inspired by recent progress in diffusion-based camera pose estimation, we embed this formulation into a diffusion transformer architecture that aligns a query image with a set of posed templates. We reparameterize object rotation using object-centered camera rays and model object translation by extending scale-invariant translation estimation to dense translation offsets. Our model leverages geometric priors from the templates to guide accurate query pose inference. A coarse-to-fine training strategy based on narrowed template sampling improves performance without modifying the network architecture. Extensive experiments across multiple benchmark datasets show competitive results of our method compared to state-of-the-art approaches in unseen object pose estimation.

Problem

• Improving 6D object pose estimation for unseen objects
• Reformulating pose estimation as ray alignment problem
• Leveraging geometric priors from posed template images

Innovation

• Reformulates pose estimation as ray alignment problem
• Uses diffusion transformer for query-template alignment
• Models translation via dense scale-invariant offsets