cs.CVOct 24, 2025

Epipolar Geometry Improves Video Generation Models

Authors: Orest Kupyn, Théo Uscidda, Marta Tintore Gazulla, Fabian Manhardt, Federico Tombari, Christian Rupprecht

Organizations: University of Oxford · Google Research · CREST-ENSAE, Institut Polytechnique de Paris · Technical University of Munich

Abstract

Video generation models have advanced significantly through the latent diffusion transformers trained with rectified flow techniques. Yet these models still struggle with geometric inconsistencies, unstable motion, and visual artifacts that break the illusion of realistic 3D scenes. 3D-consistent video generation could significantly impact numerous downstream applications in generation and reconstruction tasks. We explore how epipolar geometry constraints improve modern video diffusion models. Despite using massive training data, these models fail to capture fundamental geometric principles. We align diffusion models using pairwise epipolar geometry constraints via preference-based optimization, directly addressing unstable trajectories and geometric artifacts through mathematically principled geometric enforcement. Our approach efficiently enforces geometric principles without requiring end-to-end differentiability. Evaluation demonstrates that classical geometric constraints provide more reliable preference signals than the video-quality and geometry metrics based on foundational models. We construct preferences from static scenes with camera motion, where the epipolar metric is most reliable, while the aligned model generalizes to dynamic scenes. By bridging data-driven learning with classical computer vision, we reduce epipolar error by 31% and improve human-rated consistency from 54% to 72% without compromising visual quality.

Figures & tables

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. GeoFlow: Enforcing Implicit Geometric Consistency in Video Generation

    May 18, 2026Jan Ackermann, Shengqu Cai, Boyang Deng +3Physical Consistency in Video GenerationVideo Diffusion Models

  2. VideoWeave: Unlocking Geometric Consistency in Video Generation via Joint Geometry-Video Modeling

    Jun 12, 2026Xunzhi Xiang, Zixuan Duan, Yabo Chen +8Physical Consistency in Video GenerationImage-to-Video Generation

  3. VIGOR: VIdeo Geometry-Oriented Reward for Temporal Generative Alignment

    Mar 17, 2026Tengjiao Yin, Jinglei Shi, Heng Guo +1Diffusion Model AlignmentMulti-View Consistency