4Real-Video: Learning Generalizable Photo-Realistic 4D Video Diffusion

We propose 4Real-Video, a novel framework for generating 4D videos, organized as a grid of video frames with both time and viewpoint axes. In this grid, each row contains frames sharing the same timestep, while each column contains frames from the same viewpoint. We propose a novel two-stream archit...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-12
Main Authors: Wang, Chaoyang, Zhuang, Peiye, Ngo, Tuan Duc, Menapace, Willi, Siarohin, Aliaksandr, Vasilkovsky, Michael, Skorokhodov, Ivan, Tulyakov, Sergey, Wonka, Peter, Hsin-Ying, Lee
Format: Article
Language:English
Subjects:
Diffusion layers
Diffusion rate
Frames (data processing)
Synchronism
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose 4Real-Video, a novel framework for generating 4D videos, organized as a grid of video frames with both time and viewpoint axes. In this grid, each row contains frames sharing the same timestep, while each column contains frames from the same viewpoint. We propose a novel two-stream architecture. One stream performs viewpoint updates on columns, and the other stream performs temporal updates on rows. After each diffusion transformer layer, a synchronization layer exchanges information between the two token streams. We propose two implementations of the synchronization layer, using either hard or soft synchronization. This feedforward architecture improves upon previous work in three ways: higher inference speed, enhanced visual quality (measured by FVD, CLIP, and VideoScore), and improved temporal and viewpoint consistency (measured by VideoScore and Dust3R-Confidence).
ISSN:2331-8422