Flow of Truth: Proactive Temporal Forensics for Image-to-Video Generation

2026-04-16

AI Summary

This work addresses the challenge that existing image-to-video generation methods produce temporally evolving content, rendering conventional static-image forensic techniques ineffective at tracking pixel-level transformations over time and thus unable to reliably localize forgery traces. To overcome this limitation, we propose the first temporal proactive forensic framework tailored for image-to-video generation, modeling video synthesis not as frame-by-frame creation but as a temporally traceable motion process of pixels across time. By introducing a learnable forensic template and a template-guided optical flow module, our approach decouples motion from appearance information, enabling consistent temporal tracking of pixel evolution trajectories. Extensive experiments demonstrate that the method generalizes effectively across diverse commercial and open-source image-to-video generators and significantly improves temporal forgery detection performance.

Technology Category

• Computer Vision: Motion & Tracking
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

The rapid rise of image-to-video (I2V) generation enables realistic videos to be created from a single image but also brings new forensic demands. Unlike static images, I2V content evolves over time, requiring forensics to move beyond 2D pixel-level tampering localization toward tracing how pixels flow and transform throughout the video. As frames progress, embedded traces drift and deform, making traditional spatial forensics ineffective. To address this unexplored dimension, we present **Flow of Truth**, the first proactive framework focusing on temporal forensics in I2V generation. A key challenge lies in discovering a forensic signature that can evolve consistently with the generation process, which is inherently a creative transformation rather than a deterministic reconstruction. Despite this intrinsic difficulty, we innovatively redefine video generation as *the motion of pixels through time rather than the synthesis of frames*. Building on this view, we propose a learnable forensic template that follows pixel motion and a template-guided flow module that decouples motion from image content, enabling robust temporal tracing. Experiments show that Flow of Truth generalizes across commercial and open-source I2V models, substantially improving temporal forensics performance.

Problem

• image-to-video generation
• temporal forensics
• pixel flow
• video authenticity
• deepfake detection

Innovation

• temporal forensics
• image-to-video generation
• pixel motion
• learnable forensic template
• template-guided flow