New framework MotionPhys detects AI-generated videos via motion trajectory analysis

Researchers have developed MotionPhys, a new framework designed to detect AI-generated videos by analyzing the physical consistency of optical-flow trajectories. Unlike existing methods that focus on visual artifacts or generator-specific traces, MotionPhys models the geometric evolution of object motion over time. This approach reveals subtle physical inconsistencies that are not apparent in short sequences, making it effective across various AI video generation models. AI

IMPACT This research could lead to more robust detection of synthetic media by focusing on physical plausibility rather than visual artifacts.

RANK_REASON The cluster contains a research paper detailing a new method for detecting AI-generated videos. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• MotionPhys
• ScienceCast

• paper
• other

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