Video Latent Flow Matching: Optimal Polynomial Projections for Video Interpolation and Extrapolation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cao, Yang, Song, Zhao, Yang, Chiwun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912218787872768
author Cao, Yang
Song, Zhao
Yang, Chiwun
author_facet Cao, Yang
Song, Zhao
Yang, Chiwun
contents This paper considers an efficient video modeling process called Video Latent Flow Matching (VLFM). Unlike prior works, which randomly sampled latent patches for video generation, our method relies on current strong pre-trained image generation models, modeling a certain caption-guided flow of latent patches that can be decoded to time-dependent video frames. We first speculate multiple images of a video are differentiable with respect to time in some latent space. Based on this conjecture, we introduce the HiPPO framework to approximate the optimal projection for polynomials to generate the probability path. Our approach gains the theoretical benefits of the bounded universal approximation error and timescale robustness. Moreover, VLFM processes the interpolation and extrapolation abilities for video generation with arbitrary frame rates. We conduct experiments on several text-to-video datasets to showcase the effectiveness of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2502_00500
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Video Latent Flow Matching: Optimal Polynomial Projections for Video Interpolation and Extrapolation
Cao, Yang
Song, Zhao
Yang, Chiwun
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
This paper considers an efficient video modeling process called Video Latent Flow Matching (VLFM). Unlike prior works, which randomly sampled latent patches for video generation, our method relies on current strong pre-trained image generation models, modeling a certain caption-guided flow of latent patches that can be decoded to time-dependent video frames. We first speculate multiple images of a video are differentiable with respect to time in some latent space. Based on this conjecture, we introduce the HiPPO framework to approximate the optimal projection for polynomials to generate the probability path. Our approach gains the theoretical benefits of the bounded universal approximation error and timescale robustness. Moreover, VLFM processes the interpolation and extrapolation abilities for video generation with arbitrary frame rates. We conduct experiments on several text-to-video datasets to showcase the effectiveness of our method.
title Video Latent Flow Matching: Optimal Polynomial Projections for Video Interpolation and Extrapolation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2502.00500