Redirecting to original paper in 30 seconds...
Click below to go immediately or wait for automatic redirect
📄 Abstract
Abstract: Accurate 6D pose estimation of 3D objects is a fundamental task in computer
vision, and current research typically predicts the 6D pose by establishing
correspondences between 2D image features and 3D model features. However, these
methods often face difficulties with textureless objects and varying
illumination conditions. To overcome these limitations, we propose GS2POSE, a
novel approach for 6D object pose estimation. GS2POSE formulates a pose
regression algorithm inspired by the principles of Bundle Adjustment (BA). By
leveraging Lie algebra, we extend the capabilities of 3DGS to develop a
pose-differentiable rendering pipeline, which iteratively optimizes the pose by
comparing the input image to the rendered image. Additionally, GS2POSE updates
color parameters within the 3DGS model, enhancing its adaptability to changes
in illumination. Compared to previous models, GS2POSE demonstrates accuracy
improvements of 1.4\%, 2.8\% and 2.5\% on the T-LESS, LineMod-Occlusion and
LineMod datasets, respectively.
Authors (7)
Junbo Li
Weimin Yuan
Yinuo Wang
Yue Zeng
Shihao Shu
Cai Meng
+1 more
Submitted
October 19, 2025
Key Contributions
GS2POSE integrates Gaussian Splatting with pose estimation by formulating a Bundle Adjustment-inspired regression algorithm using Lie algebra. It enables a pose-differentiable rendering pipeline that iteratively optimizes pose and updates color parameters, leading to improved accuracy and robustness, especially for textureless objects and varying lighting conditions.
Business Value
Enhances the reliability of object recognition and manipulation for robots and AR/VR systems, enabling more precise interaction with the physical world even in challenging visual conditions.