cs.CVNov 12, 2026

Consistent Video-to-Video Translation via Explicit Correspondences

Authors: Gaurav Parmar, Zhengqi Li, Richard Zhang, Jun-Yan Zhu, Srinivasa Narasimhan, Eli Shechtman, Yotam Nitzan

Organizations: Carnegie Mellon University · Adobe Research

Abstract

Interactive video-to-video applications require real-time generation while maintaining long-range temporal consistency. However, recent methods achieve speed by restricting attention to a fixed-length causal window. This limits long-range consistency whenever relevant history falls outside the context window. To address this limitation, we introduce vid2vid-long, a correspondence-based approach grounded in a simple observation. Video-to-video translation provides an explicit cross-frame reference: the regions that match in the input video should match in the output video too. Concretely, we compute patch-level correspondences in the input, guiding retrieval of previously generated outputs in the corresponding regions, and condition the model on these patches through correspondence cross-attention. Our experiments demonstrate that vid2vid-long consistently improves long-range temporal consistency across recent video models while preserving generation quality and real-time performance.

Figures & tables

Explore similar work

CardsList
  1. MilliVid: Hierarchical Latents for Long-Range Consistency in Video Generation

    Jun 8, 2026Ishaan Preetam Chandratreya, David Charatan, Basile Van Hoorick +4Video Diffusion ModelsLong Video Generation

  2. BIFE: Better Interaction, Fewer Errors for Minute-Long Video Generation

    Nov 28, 2025Zeyu Zhang, Jinyuan Mao, Shuning Chang +5Long-Horizon Video GenerationAutoregressive Video Generation

  3. Recast the Past: Stable Long Video Generation with Full Context Memory

    Oct 20, 2026Yukuo Ma, Cong Liu, Junke Wang +4Long-Horizon Video GenerationMemory-Augmented Video Generation