GeoReF: Geometric Alignment Across Shape Variation for Category-level Object Pose Refinement
This work addresses the problem of robust object pose estimation for robotics and computer vision applications, but it is incremental as it builds on existing instance-level refinement methods.
The paper tackles category-level object pose refinement, which is challenging due to shape variations within categories, by introducing a novel architecture that integrates an HS-layer, learnable affine transformations, and a cross-cloud transformation mechanism, resulting in significant improvements over baseline methods across all metrics.
Object pose refinement is essential for robust object pose estimation. Previous work has made significant progress towards instance-level object pose refinement. Yet, category-level pose refinement is a more challenging problem due to large shape variations within a category and the discrepancies between the target object and the shape prior. To address these challenges, we introduce a novel architecture for category-level object pose refinement. Our approach integrates an HS-layer and learnable affine transformations, which aims to enhance the extraction and alignment of geometric information. Additionally, we introduce a cross-cloud transformation mechanism that efficiently merges diverse data sources. Finally, we push the limits of our model by incorporating the shape prior information for translation and size error prediction. We conducted extensive experiments to demonstrate the effectiveness of the proposed framework. Through extensive quantitative experiments, we demonstrate significant improvement over the baseline method by a large margin across all metrics.