GAP-GDRNet: Geometry-Aware Monocular Visual Pose Sensing on a Single-Target Synthetic Spacecraft Dataset
2026-07-02 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors developed GAP-GDRNet, a system to help spacecraft figure out their exact position and orientation using just a single camera image. Their method improves on previous work by adding modules that pay special attention to important features and geometric details, even when the spacecraft has weak textures or parts that are hard to see. They also created a way to generate detailed training data using Blender, a 3D software, to teach the system how to recognize different spacecraft poses accurately. Overall, their work aims to make pose estimation more reliable in challenging space images.
Monocular relative pose sensing6D pose estimationGeometry-guided direct regressionAttention mechanismsRGB imagesPatch-PnP algorithmFeature refinementBlender annotationOn-orbit servicingSparse geometric evidence
Authors
Yonglong Zhang, Yang Liu
Abstract
Monocular relative pose sensing is a central perception problem in non-cooperative rendezvous and on-orbit servicing. In spacecraft images, however, weak surface texture, thin appendages, illumination changes, and partial occlusion often leave only sparse and unstable geometric evidence. This article presents GAP-GDRNet, a geometry-aware attention-enhanced framework for monocular RGB-based 6D pose sensing. The method follows the geometry-guided direct regression paradigm of GDR-Net and modifies two points in the pipeline: an attention-based feature refinement (AFR) module is placed before dense geometric prediction, and a patch-level geometric self-attention (PGSA) module is inserted into Patch-PnP. AFR reinforces global spacecraft structure together with local weak-texture cues; PGSA then relates downsampled geometric patches before final pose regression. A Blender-based annotation process supplies target masks, visible-region masks, dense model-coordinate maps, camera intrinsics, and 6D pose labels for supervised training.