ASTRA framework uses 2D motion trajectories for asynchronous 3D scene reconstruction

Researchers have developed ASTRA, a novel framework for reconstructing dynamic 3D scenes from asynchronous multi-camera inputs. Unlike previous methods that rely on photometric supervision, ASTRA utilizes 2D motion trajectories as explicit, texture-agnostic guidance. This approach improves robustness against temporal asynchrony, even with offsets up to 25 frames, by aligning projected 3D point motion with observed 2D trajectories. The framework demonstrates significant improvements in PSNR, reduces temporal-offset mean absolute error, and substantially increases synchronization success rates. AI

IMPACT This research could improve the accuracy and robustness of 3D reconstruction in real-world applications where camera synchronization is challenging.

RANK_REASON The item is an academic paper detailing a new method for 3D scene reconstruction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• ASTRA
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• dynamic Gaussian splatting
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →