MPF-Net: Exposing High-Fidelity AI-Generated Video Forgeries via Hierarchical Manifold Deviation and Micro-Temporal Fluctuations

2026-01-29

AI Summary

Current high-fidelity AI-generated videos exhibit near-imperceptible artifacts in macroscopic semantics and temporal consistency, necessitating deeper investigation into intrinsic generative mechanism discrepancies. This work proposes a hierarchical dual-path detection framework: one branch leverages vision foundation models to identify spatial anomalies that deviate from the authentic data manifold, while the other introduces the novel concept of “manifold projection fluctuation,” employing a micro-temporal fluctuation module to capture structured homogeneity patterns in inter-frame residuals of AI-generated videos. By cascading these complementary pathways, the method jointly exploits macro-level manifold deviations and micro-level computational fingerprints, effectively circumventing the limitations of conventional spatiotemporal cues. The approach achieves significantly improved detection accuracy on state-of-the-art generative models such as Veo and Wan, demonstrating robustness against highly realistic synthetic videos.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Other Foundations of Human Computation & AI
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

With the rapid advancement of video generation models such as Veo and Wan, the visual quality of synthetic content has reached a level where macro-level semantic errors and temporal inconsistencies are no longer prominent. However, this does not imply that the distinction between real and cutting-edge high-fidelity fake is untraceable. We argue that AI-generated videos are essentially products of a manifold-fitting process rather than a physical recording. Consequently, the pixel composition logic of consecutive adjacent frames residual in AI videos exhibits a structured and homogenous characteristic. We term this phenomenon `Manifold Projection Fluctuations'(MPF). Driven by this insight, we propose a hierarchical dual-path framework that operates as a sequential filtering process. The first, the Static Manifold Deviation Branch, leverages the refined perceptual boundaries of Large-Scale Vision Foundation Models (VFMs) to capture residual spatial anomalies or physical violations that deviate from the natural real-world manifold (off-manifold). For the remaining high-fidelity videos that successfully reside on-manifold and evade spatial detection, we introduce the Micro-Temporal Fluctuation Branch as a secondary, fine-grained filter. By analyzing the structured MPF that persists even in visually perfect sequences, our framework ensures that forgeries are exposed regardless of whether they manifest as global real-world manifold deviations or subtle computational fingerprints.

Problem

• AI-generated video
• video forgery detection
• manifold deviation
• temporal fluctuations
• high-fidelity synthesis

Innovation

• Manifold Projection Fluctuations
• Hierarchical Dual-Path Framework
• Micro-Temporal Fluctuations
• AI-Generated Video Detection
• Vision Foundation Models