TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: YU, Mark, Hu, Wenbo, Xing, Jinbo, Shan, Ying
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912264732278784
author YU, Mark
Hu, Wenbo
Xing, Jinbo
Shan, Ying
author_facet YU, Mark
Hu, Wenbo
Xing, Jinbo
Shan, Ying
contents We present TrajectoryCrafter, a novel approach to redirect camera trajectories for monocular videos. By disentangling deterministic view transformations from stochastic content generation, our method achieves precise control over user-specified camera trajectories. We propose a novel dual-stream conditional video diffusion model that concurrently integrates point cloud renders and source videos as conditions, ensuring accurate view transformations and coherent 4D content generation. Instead of leveraging scarce multi-view videos, we curate a hybrid training dataset combining web-scale monocular videos with static multi-view datasets, by our innovative double-reprojection strategy, significantly fostering robust generalization across diverse scenes. Extensive evaluations on multi-view and large-scale monocular videos demonstrate the superior performance of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2503_05638
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models
YU, Mark
Hu, Wenbo
Xing, Jinbo
Shan, Ying
Computer Vision and Pattern Recognition
Artificial Intelligence
Graphics
We present TrajectoryCrafter, a novel approach to redirect camera trajectories for monocular videos. By disentangling deterministic view transformations from stochastic content generation, our method achieves precise control over user-specified camera trajectories. We propose a novel dual-stream conditional video diffusion model that concurrently integrates point cloud renders and source videos as conditions, ensuring accurate view transformations and coherent 4D content generation. Instead of leveraging scarce multi-view videos, we curate a hybrid training dataset combining web-scale monocular videos with static multi-view datasets, by our innovative double-reprojection strategy, significantly fostering robust generalization across diverse scenes. Extensive evaluations on multi-view and large-scale monocular videos demonstrate the superior performance of our method.
title TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Graphics
url https://arxiv.org/abs/2503.05638