Video Prediction by Modeling Videos as Continuous Multi-Dimensional Processes

Diffusion models have made significant strides in image generation, mastering tasks such as unconditional image synthesis, text-image translation, and image-to-image conversions. However, their capability falls short in the realm of video prediction, mainly because they treat videos as a collection...

Full description

Saved in:
Bibliographic Details
Main Authors: Shrivastava, Gaurav, Shrivastava, Abhinav
Format: Conference Proceeding
Language:English
Subjects:
Benchmark testing
Coherence
Computer vision
Diffusion Models
Frame-conditional video generation
Image synthesis
Predictive models
Technological innovation
Training
Video Prediction
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Diffusion models have made significant strides in image generation, mastering tasks such as unconditional image synthesis, text-image translation, and image-to-image conversions. However, their capability falls short in the realm of video prediction, mainly because they treat videos as a collection of independent images, relying on external constraints such as temporal attention mechanisms to enforce temporal coherence. In our paper, we introduce a novel model class, that treats video as a continuous multi-dimensional process rather than a series of discrete frames. Through extensive ex-perimentation, we establish state-of-the-art performance in video prediction, validated on benchmark datasets including KTH, BAIR, Human3.6M, and UCF101 1 1 Navigate to the webpage for video results.
ISSN:2575-7075
DOI:10.1109/CVPR52733.2024.00691