Generation Is Compression: Zero-Shot Video Coding via Stochastic Rectified Flow

📰 ArXiv cs.AI

Generative Video Codec (GVC) uses a pretrained video generative model as a zero-shot codec, eliminating the need for retraining

advanced Published 30 Mar 2026
Action Steps
  1. Convert deterministic rectified-flow ODE to stochastic rectified flow
  2. Use pretrained video generative model as the codec itself
  3. Specify generative decoding trajectory via transmitted bitstream
  4. Enable zero-shot video coding without retraining
Who Needs to Know This

This benefits AI engineers and researchers working on video compression and generative models, as it provides a novel approach to video coding

Key Insight

💡 Generative models can be used as codecs themselves, eliminating the need for retraining

Share This
💡 Zero-shot video coding via stochastic rectified flow!
Read full paper → ← Back to News