4DVD: Cascaded Dense-view Video Diffusion Model for High-quality 4D Content Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Shuzhou, Cun, Xiaodong, Li, Xiaoyu, Li, Yaowei, Zhang, Jian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913977679740928
author Yang, Shuzhou
Cun, Xiaodong
Li, Xiaoyu
Li, Yaowei
Zhang, Jian
author_facet Yang, Shuzhou
Cun, Xiaodong
Li, Xiaoyu
Li, Yaowei
Zhang, Jian
contents Given the high complexity of directly generating high-dimensional data such as 4D, we present 4DVD, a cascaded video diffusion model that generates 4D content in a decoupled manner. Unlike previous multi-view video methods that directly model 3D space and temporal features simultaneously with stacked cross view/temporal attention modules, 4DVD decouples this into two subtasks: coarse multi-view layout generation and structure-aware conditional generation, and effectively unifies them. Specifically, given a monocular video, 4DVD first predicts the dense view content of its layout with superior cross-view and temporal consistency. Based on the produced layout priors, a structure-aware spatio-temporal generation branch is developed, combining these coarse structural priors with the exquisite appearance content of input monocular video to generate final high-quality dense-view videos. Benefit from this, explicit 4D representation~(such as 4D Gaussian) can be optimized accurately, enabling wider practical application. To train 4DVD, we collect a dynamic 3D object dataset, called D-Objaverse, from the Objaverse benchmark and render 16 videos with 21 frames for each object. Extensive experiments demonstrate our state-of-the-art performance on both novel view synthesis and 4D generation. Our project page is https://4dvd.github.io/
format Preprint
id arxiv_https___arxiv_org_abs_2508_04467
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle 4DVD: Cascaded Dense-view Video Diffusion Model for High-quality 4D Content Generation
Yang, Shuzhou
Cun, Xiaodong
Li, Xiaoyu
Li, Yaowei
Zhang, Jian
Computer Vision and Pattern Recognition
Given the high complexity of directly generating high-dimensional data such as 4D, we present 4DVD, a cascaded video diffusion model that generates 4D content in a decoupled manner. Unlike previous multi-view video methods that directly model 3D space and temporal features simultaneously with stacked cross view/temporal attention modules, 4DVD decouples this into two subtasks: coarse multi-view layout generation and structure-aware conditional generation, and effectively unifies them. Specifically, given a monocular video, 4DVD first predicts the dense view content of its layout with superior cross-view and temporal consistency. Based on the produced layout priors, a structure-aware spatio-temporal generation branch is developed, combining these coarse structural priors with the exquisite appearance content of input monocular video to generate final high-quality dense-view videos. Benefit from this, explicit 4D representation~(such as 4D Gaussian) can be optimized accurately, enabling wider practical application. To train 4DVD, we collect a dynamic 3D object dataset, called D-Objaverse, from the Objaverse benchmark and render 16 videos with 21 frames for each object. Extensive experiments demonstrate our state-of-the-art performance on both novel view synthesis and 4D generation. Our project page is https://4dvd.github.io/
title 4DVD: Cascaded Dense-view Video Diffusion Model for High-quality 4D Content Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.04467