Space & aviationResearch findings
Lighting-Disentangled Vision Transformer for Robust Spacecraft 6D Pose Estimation
Original published 2026.06.10 12:30 KSTAdded to Blackdeck 2026.11.22 12:05 KST
What has changed?
Separates illumination-sensitive appearance tokens from pose-relevant structure tokens, enabling reliable pose estimation under varying lighting conditions
Key points
- Proposes a model that improves robustness to illumination variation and specular reflection in spacecraft pose estimation
- Disentangles illumination-sensitive appearance tokens from pose-relevant structure tokens to assess visual reliability
- Demonstrates reduced translation error and enhanced stability on the SPEED+ V2 benchmark under sunlamp conditions
Connect it to your work
Maintains stable pose estimation despite specular reflection and shadowing in challenging lighting environments
What remains unproven
This research note is based on an abstract; check the original for test conditions and application scope.
Originals used for this note
Links open the original publisher in a new tab.