ScaleVid framework enables geometry-aware video object scaling without 3D reconstruction

Researchers have developed ScaleVid, a novel framework for geometry-aware video object scaling that aims to resize objects anisotropically while maintaining geometric plausibility and temporal coherence. This method avoids costly 3D reconstruction by employing a progressive two-stage training process that separates foreground transformation from background preservation. The framework synthesizes realistic video compositions without requiring paired real-world scaling targets, demonstrating superior geometric consistency and faster inference compared to existing approaches. AI

IMPACT Enables more precise and efficient manipulation of video content by allowing objects to be resized while maintaining visual integrity.

RANK_REASON The item is a research paper detailing a new method for video object scaling. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScaleVid
• ScienceCast
• scite Smart Citations

• paper
• other

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